# Flyte Open Source Documentation > Full documentation (single file): https://www.union.ai/docs/v2/flyte/llms-full.txt > Site: https://www.union.ai/docs/v2/flyte Each entry below is `- [Page title](URL)` followed by the H2/H3 headings found on that page. Each page's markdown is served at its own URL with `.md` appended, e.g. `/integrations/hydra.md`. A section landing page's markdown also lists every page directly beneath it, each with a one-line description, so it is the index for that section. ## User guide - [Get started](https://www.union.ai/docs/v2/flyte/user-guide/get-started.md) - Pure Python, no DSL - Durability - Reproducibility - Recoverability - Built for scale - What this means in practice - Start here - [Quickstart](quickstart.md) - [Core concepts](core-concepts.md) - [Run modes](run-modes.md) - [Get started > Quickstart](https://www.union.ai/docs/v2/flyte/user-guide/get-started/quickstart.md) - What you'll need - Install the SDK - Run something straight away - Configure - Write your first workflow - Run it - See the results - Next steps - [Get started > Core concepts](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts.md) - How a Flyte program fits together - [Get started > Core concepts > TaskEnvironment](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts/task-environment.md) - A minimal example - What TaskEnvironment controls - Configuring resources - Configuring container images - Multiple tasks, one environment - Multiple environments - Next steps - [Get started > Core concepts > Tasks](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts/tasks.md) - Defining a task - Type hints are required - Tasks calling tasks - The top-level task - Running tasks locally - Running tasks remotely - Next steps - [Get started > Core concepts > Runs and actions](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts/runs-and-actions.md) - What is a run? - What is an action? - Runs vs actions in practice - Viewing runs in the UI - Understanding the execution graph - Checking run status - Next steps - [Get started > Core concepts > Where your data lives](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts/where-data-lives.md) - The two stores - What goes in the database - What goes in the bucket - What "literal" and "raw data" mean - What "metadata" means - 1. "Metadata" as in the control plane database (Flyte's usage) - 2. "Metadata bucket" (a deployment/ops term you may see) - Per-run customization: `raw_data_path` - What happens if the bucket is purged - The short version - [Get started > Core concepts > Apps](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts/introducing-apps.md) - Tasks vs apps - AppEnvironment - A hello world app - Understanding the code - Serving the app - When to use apps vs tasks - Common patterns - Next steps - [Get started > Core concepts > Projects and domains](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts/projects-and-domains.md) - How projects and domains are used - Managing projects via CLI - Create a project - List projects - Update a project - Archive a project - Unarchive a project - Listing projects programmatically - Managing projects via the UI - Domains - Targeting a domain - [Get started > Core concepts > Key capabilities](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts/key-capabilities.md) - Environment and resources - Deployment - Data handling - Parallelism and composition - Security and automation - Durability and reliability - Apps and serving - Notebooks - Next steps - [Get started > Core concepts > Basic project: RAG](https://www.union.ai/docs/v2/flyte/user-guide/get-started/core-concepts/basic-project.md) - Concepts covered - Part 1: The embedding pipeline - Setting up the environment - Fetching data - Creating embeddings - Orchestrating the pipeline - Running the pipeline - Part 2: The serving application - App environment configuration - The Streamlit application - Deploying the app - Key takeaways - [Get started > Run modes](https://www.union.ai/docs/v2/flyte/user-guide/get-started/run-modes.md) - Local and remote - [Local](running-locally.md) - [Devbox](running-devbox.md) - [Deployed cluster](running-remote.md) - [Get started > Run modes > Run locally in Python](https://www.union.ai/docs/v2/flyte/user-guide/get-started/run-modes/running-locally.md) - Getting started - Running tasks locally - Two ways to run a task - As a script: `python hello.py` - Via the CLI: `flyte run` - Terminal UI - Exploring past runs - What works locally - From in-process to on-cluster - Next steps - [Get started > Run modes > Run on the Devbox](https://www.union.ai/docs/v2/flyte/user-guide/get-started/run-modes/running-devbox.md) - What you'll need - Install the SDK - Start the devbox - CPU - GPU - Check the devbox status - Configure - Run a workflow on the devbox - View results in the UI - Stop the devbox - Inline configuration - Programmatic - CLI - Delete the devbox - Using a CUDA-enabled GPU host - Next steps - [Get started > Run modes > Run on a deployed cluster](https://www.union.ai/docs/v2/flyte/user-guide/get-started/run-modes/running-remote.md) - Prerequisites - Install the flyte package - Configuration file - Set up local Docker - Using the configuration - Explicit configuration - Programmatic - CLI - Programmatic - CLI - Check current configuration - Inline configuration - Programmatic - CLI - Next steps - [Tasks](https://www.union.ai/docs/v2/flyte/user-guide/tasks.md) - [Configure tasks](task-configuration.md) - [Build tasks](task-programming.md) - [Run and deploy tasks](task-deployment.md) - Related - [Apps](../apps.md) - [Agents](../agents.md) - [Tasks > Configure tasks](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration.md) - Task configuration levels - Example - Task configuration parameters - Scheduled and reactive automation - Distributed tasks with a clustered environment - [Tasks > Configure tasks > Container images](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/container-images.md) - Specifying your own image directly - Specifying your own image with the `flyte.Image` object - Example: Defining a custom image with `Image.from_debian_base` - Example: Defining an image based on uv script metadata - Customizing an image with `with_*` methods - Installing dependencies from a `uv` or Poetry project - Installing dependencies from a pixi project - Copying local files into the image - More ways to define a base image - Starting from an existing image - Building from a Dockerfile - Referencing an image defined in configuration - Using different images for different tasks - Image building - Configuring the `builder` - Local image building - Remote `ImageBuilder` - Install private PyPI packages - [Tasks > Configure tasks > Resources](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/resources.md) - Resources data class - Ephemeral storage - Examples - Usage in TaskEnvironment - Usage in a task-specific override - Accelerators - NVIDIA GPUs - Google TPUs - AMD GPUs - AWS Trainium and Inferentia (Neuron) - Intel Habana Gaudi - [Tasks > Configure tasks > Secrets](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/secrets.md) - Creating a literal string secret - Creating a file secret - Scoping secrets - Listing secrets - Deleting secrets - Declaring secrets in a task environment - Using a literal string secret - Using a file secret - Using secrets in local runs - Environment variable secrets - File secrets - Overriding secrets at invocation time - [Tasks > Configure tasks > Caching](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/caching.md) - Overview - Basic caching usage - `"auto"` - automatic versioning - `"override"` - `"disable"` - No caching - Advanced caching configuration - Ignoring specific inputs - Cache serialization - Salt for cache key variation - Caching with `flyte.map` - Excluding a bound argument from the per-element key - Content-based caching for DataFrames, files, and directories - DataFrames - Files - Directories - Cache policies - Function body policy (default) - Custom cache policies - Caching configuration at different levels - `TaskEnvironment` level - `@env.task` decorator level - `task.override` level - Runtime cache control - Project and domain cache isolation - Local development caching - [Tasks > Configure tasks > Reusable containers](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/reusable-containers.md) - [Tasks > Configure tasks > Pod templates](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/pod-templates.md) - How it works - Requirements - Basic usage - PodTemplate parameters - Volume mounts - GCS/S3 volume mounts - Sidecar containers - Image pull secrets - Cluster-specific configuration - Important notes - Best practices - Learn more - [Tasks > Configure tasks > Multiple environments](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/multiple-environments.md) - Constraints on multiple environments - Task `depends_on` constraints - Dependency inclusion constraints - Example - [Tasks > Configure tasks > Clustered task environment](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/clustered-task-environment.md) - When to use it - How it works - Prerequisites - Basic usage - Configuration parameters - `TorchRun` - `ClusterFailurePolicy` - The distributed context - Checkpointing - Using higher-level frameworks - Constraints - [Tasks > Configure tasks > Retries and timeouts](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/retries-and-timeouts.md) - The action lifecycle - Retries - Retry count - Retries with exponential backoff - Skip retries for failures that can't be fixed - System retries - Timeouts - `max_runtime`: bound a single attempt's execution - `max_queued_time`: fail fast when capacity isn't available - `deadline`: bound the total wall-clock - Combining the bounds - Combining retries and timeouts - [Tasks > Configure tasks > Interruptible tasks](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/interruptible-tasks-and-queues.md) - Setting at different levels - Behavior on preemption - Fall back to on-demand when spot capacity is unavailable - [Tasks > Configure tasks > Task plugins](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/task-plugins.md) - Default execution: containers - Compute plugins - Available compute plugins - How compute plugins work - Using compute plugins - Using plugins on Union - Backend integrations - Next steps - [Tasks > Configure tasks > Additional task settings](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/additional-task-settings.md) - Naming and metadata - `name` - `short_name` - `description` - `docs` - `report` - Source-code link (automatic) - `links` - Default inputs - Environment variables - Inline I/O threshold - [Tasks > Configure tasks > Logging](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/logging.md) - Set the logging level with environment variables - Set the framework verbosity from the CLI - Set the logging level with `flyte.init()` - Write to the user logger - JSON logging and third-party libraries - Set it at registration with `FLYTE_RESET_ROOT_LOGGER` - Related - [Tasks > Configure tasks > Overrides](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-configuration/overrides.md) - What you can override - What you cannot override - Overriding when a task is reusable - Overriding secrets - [Tasks > Build tasks](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming.md) - What you'll learn - When to use these patterns - [Tasks > Build tasks > Files and directories](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/files-and-directories.md) - Example usage - Synchronous and asynchronous APIs - File methods - Dir methods - Synchronous example - JSONL files - [Tasks > Build tasks > Data classes and structures](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/dataclasses-and-structures.md) - Example: Combining dataclasses and Pydantic models - [Tasks > Build tasks > DataFrames](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/dataframes.md) - Setting up the environment and sample data - Create a raw DataFrame - Create a flyte.io.DataFrame - Automatically convert between types - Downloading DataFrames - Run the example - Polars DataFrames - Setup - Eager DataFrames - Lazy DataFrames - Run the example - Lance datasets - See also - [Tasks > Build tasks > Custom types](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/handling-custom-types.md) - Types of extensions - Creating a type transformer - Step 1: Define your custom type - Step 2: Create the type transformer - Step 3: Register the transformer - Distributing type plugins - Configure pyproject.toml - Automatic loading - Controlling plugin loading - Using custom types in tasks - DataFrame extensions - Best practices - See also - [Tasks > Build tasks > Custom context](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/custom-context.md) - Overview - When to use it and when not to - Setting custom context - Run-level context - Overriding inside a task (local override that affects nested tasks) - Adding new keys for nested tasks - Accessing custom context - [Tasks > Build tasks > Abort and cancel actions](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/abort-tasks.md) - Action lifetime - Canceling actions programmatically - External abort - Aborting via the CLI - Handling external aborts - [Tasks > Build tasks > Run a bioinformatics tool](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/container-tasks.md) - What are container tasks? - How data flows in and out - Basic usage - Template syntax for inputs - Using container tasks in workflows - Advanced: Passing files and directories - Use case: Agentic sandbox execution - Use case: Legacy and specialized containers - Use case: Multi-language workflows - Configuration options - ContainerTask parameters - Supported input/output types - Best practices - Local execution - When to use container tasks - [Tasks > Build tasks > Links](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/links.md) - Creating a link - Using execution metadata - Dynamic links with override - [Tasks > Build tasks > Reports](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/reports.md) - A simple example - A more complex example - Streaming example - Rendering a custom type - Fetching a report after the run - [Tasks > Build tasks > Notebooks](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/notebooks.md) - Iterating on and running a workflow - Accessing runs and downloading logs - [Tasks > Build tasks > Remote tasks](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/remote-tasks.md) - Prerequisites - Basic usage - Understanding lazy loading - When tasks are fetched - Benefits of lazy loading - Error handling - Eager fetching with `fetch()` - Module-level vs dynamic loading - Complete example - Team A: Spark environment - Team B: ML environment - Team C: Orchestration - Invoke remote tasks in a script. - Why use remote tasks? - When to use remote tasks - How remote tasks work - Security model - Type system - Versioning options - Customizing remote tasks - Available overrides - Override examples - Chain overrides - Best practices - 1. Use meaningful task names - 2. Document task interfaces - 3. Prefer module-level loading - 4. Handle versioning thoughtfully - 5. Deploy remote tasks first - Limitations - Next steps - [Tasks > Build tasks > Error handling](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/error-handling.md) - How Flyte represents failures - Catching and recovering from errors - Out-of-memory on the GPU - Falling back when GPU capacity isn't available - Telling the timeout bounds apart - Handling GPU hardware faults (Xid errors) - Handling the failure of a run you launched - Limiting inline I/O - Natively-supported exceptions - Related pages - [Tasks > Build tasks > Traces](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/traces.md) - What are traced functions for? - What gets traced - Errors are not recorded - Supported function types - Task orchestration pattern - Relationship to caching and checkpointing - How they work together - Execution flow - Error handling and observability - Examples in practice - LLM pipeline with traces - [Tasks > Build tasks > Intra-task checkpoints](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/intra-task-checkpoints.md) - The checkpoint object - Basic usage - Async - Sync - Checkpointing ML training frameworks - PyTorch - PyTorch Lightning - Hugging Face Trainer - scikit-learn - Unsloth - How checkpoints are stored - [Tasks > Build tasks > Grouping actions](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/grouping-actions.md) - What are groups? - The problem groups solve - How groups work - Common grouping patterns - Sequential operations - Parallel processing with groups - Multi-phase workflows - Conditional grouping - Key insights - [Tasks > Build tasks > Fanout](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/fanout.md) - Understanding fanout - Example - Parallel execution - Running the example - How Flyte handles concurrency and parallelism - Iterative fanout: recursive feature elimination - [Tasks > Build tasks > Mapping over inputs](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/map.md) - Minimal example - Mapping from an async task: `flyte.map.aio` - Signature and parameters - Limiting concurrency - Handling errors - Binding constant arguments with `functools.partial` - Caching - When to use `flyte.map` - [Tasks > Build tasks > Consuming a message queue](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/consuming-a-message-queue.md) - The consumer - Define the task environment - Process a single message - The consumer loop - Run it - The producer - Notes and gotchas - [Tasks > Build tasks > Controlling parallel execution](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/controlling-parallelism.md) - The problem: unbounded parallelism - Using asyncio.Semaphore - Using flyte.map with concurrency - Running the example - When to use each approach - Per-map concurrency vs. the run-level action cap - [Tasks > Build tasks > Streaming map-reduce](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/streaming-map-reduce.md) - When to use it - Example - The driver task - Running the example - How it works - [Tasks > Build tasks > Structured concurrency with anyio](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/structured-concurrency-anyio.md) - The task-group pattern - When to use anyio - [Tasks > Build tasks > Task dependencies and ordering](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/task-dependencies.md) - The dependency graph is implicit - Ordering without a data dependency - Fan-out and fan-in - Dependency-driven scheduling - When to reach for `as_completed` - Summary - [Tasks > Build tasks > External conditions](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/conditions.md) - Supported types - Example: human approval - Example: string input at runtime - Parameters - Signaling a condition - From the CLI - From Python (remote) - Timeout with a fallback - Errors - [Tasks > Build tasks > Test business logic directly](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/unit-testing.md) - Understanding task invocation - Direct function invocation - Using `flyte.run()` - Testing business logic - Testing async tasks - Testing nested tasks - Testing type transformations and serialization - Testing nested tasks with serialization - Testing traced functions - Best practices - Quick reference - Example test suite - Future improvements - [Tasks > Build tasks > Higher-order functions](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/higher-order-functions.md) - Fallback runner - Retry with increasing memory (OOM retrier) - Circuit breaker - Auto batcher - Composing the patterns - [Tasks > Build tasks > Regular async function (not a task)](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-programming/other-features.md) - Task forwarding - Passing tasks and functions as arguments - Custom action names - Set at task definition - Override at call time - Invoking async functions from sync tasks - Async and sync task interoperability - Calling sync tasks from async tasks - Using with `flyte.map.aio()` - Using AnyIO in async tasks - [Tasks > Run and deploy tasks](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment.md) - Ephemeral deployment and immediate execution - Programmatic - CLI - Persistent deployment - Programmatic - CLI - Running already deployed tasks - Programmatic - CLI - Configuring runs with `flyte.with_runcontext()` - [Tasks > Run and deploy tasks > How task run works](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/how-task-run-works.md) - Ephemeral deployment + run: The development shortcut - Programmatic - CLI - Running deployed tasks - Programmatic - CLI - Local execution - Programmatic - CLI - Running tasks through the Union UI - Accessing task execution in the Union UI - Execution flow and architecture - Fast registration architecture - Ephemeral preparation logic - Launch methods compared - [Tasks > Run and deploy tasks > Interact with runs and actions](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/interacting-with-runs.md) - Understanding runs and actions - Key concepts - Working with runs - Retrieving a run - Programmatic - CLI - Filtering runs by label - CLI - Programmatic - UI - Watching run progress - Getting detailed run information - Working with actions - Retrieving an action - Programmatic - CLI - Nested actions - Getting detailed action information - Accessing logs - Retrieving inputs and outputs - Programmatic - CLI - Handling failures - Understanding data storage - Accessing large data from cloud storage - S3 storage access - GCS storage access - Azure Blob Storage access - Complete example - API reference - Key classes - CLI commands - Storage configuration - [Tasks > Run and deploy tasks > View logs](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/view-logs.md) - Stream logs with the CLI - Raw vs. pretty output - View the logs for a specific attempt - Filter out system logs - Scope to a project and domain - Command options - View logs in the console - Related - [Tasks > Run and deploy tasks > Work with local data](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/work-with-local-data.md) - Local execution - Uploading local data to remote runs - Uploading DataFrames - Uploading files - Uploading directories - Passing outputs between runs - Performance considerations - Summary - [Tasks > Run and deploy tasks > Run command options](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/run-command-options.md) - `--project`, `--domain` - `--run-project`, `--run-domain` - `--local` - When to use local execution - `--copy-style` - Copy style options - `--root-dir` - `--raw-data-path` - Use cases - `--service-account` - Use cases - `--name` - Benefits of custom names - `--label` - `--follow` - Behavior - `--image` - Image mapping formats - `--no-sync-local-sys-paths` - Task argument passing - Passing inputs by type - Boolean inputs - Datetime and duration inputs - Enum inputs - List, dict, dataclass, and other structured inputs - File, directory, and dataframe inputs - Optional inputs - Positional arguments - SDK options - [Tasks > Run and deploy tasks > Build a custom CLI](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/custom-cli.md) - The pattern - Example: a typed CLI with `tyro` - Using a different parser - When to use which - [Tasks > Run and deploy tasks > How task deployment works](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/how-task-deployment-works.md) - 1. Module loading and task environment discovery - Single file (default) - `--all` option - `--recursive` option - 2. Task analysis and serialization - 3. Task environment dependency resolution - 4. Code bundle creation and upload - `--copy_style loaded_modules` (default) - `--copy_style all` - `--copy_style none` - `--root-dir` option - 5. Image building - Local image building - Remote image building - 6. Source-code link discovery - How the link is built - Conditions - Understanding option relationships - [Tasks > Run and deploy tasks > Deploy command options](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/deploy-command-options.md) - `--project`, `--domain` - `--version` - When versions are used - `--dry-run` - `--all` and `--recursive` - `--copy-style` - `--copy-style loaded_modules` (default) - `--copy-style all` - `--copy-style none` - `--root-dir` - Default behavior (without `--root-dir`) - Common use cases - How it works - Example with complex project structure - `--image` - Named image mappings - Default image mapping - How it works - `--ignore-load-errors` - `--no-sync-local-sys-paths` - Default behavior (path synchronization enabled) - When to disable path synchronization - Use cases for disabling - How it works - SDK deployment options - [Tasks > Run and deploy tasks > Code packaging for on-cluster execution](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/packaging.md) - Quick comparison - Code bundling - How it works - Automatic code bundling - Manual code bundling - Including additional files with `include` - Controlling the root directory - Code bundling examples - When to use code bundling - Container-based deployment - How it works - Configuration - Image source copying methods - Complete container-based example - Using externally built images - Container-based best practices - When to use container-based deployment - Choosing the right approach - Decision tree - Hybrid approach - Troubleshooting - Import errors - Code changes not reflected - Files missing in container - Container build failures - Version conflicts - [Tasks > Run and deploy tasks > Deployment patterns](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/deployment-patterns.md) - Overview of deployment patterns - Simple file deployment - Example structure - Deployment commands - When to use - Custom Dockerfile deployment - Example structure - Alternative: Dockerfile in different directory - Key considerations - When to use - PyProject package deployment - Example structure - Business logic modules - Flyte orchestration layer - Entrypoint configuration - Dependencies and configuration - Key features - Key learning points - Usage patterns - What this example demonstrates - When to use - Package structure deployment - Example structure - Key concepts - Running with root directory - How `--root-dir` works - Alternative: Using a Python project - When to use - Full build deployment - Overview - Key configuration - Local dependency example - Critical configuration components - Configuration options - Version management best practices - Performance considerations - When to use - Troubleshooting - Python path deployment - Example structure - Implementation - Task environment dependencies - Key considerations - CLI vs direct Python execution - Best practices - Common pitfalls - When to use - Dynamic environment deployment - Domain-based environment selection - Why this pattern works - How it works - Important constraints - Alternative: environment variable approach - Usage patterns - When to use dynamic environments - Dependency and version management - Recommended: use a lockfile - No lockfile? Install against Flyte's published constraints - Choosing an approach - Best practices - Project organization - Image management - Configuration management - Development workflow - Choosing the right pattern - [Tasks > Run and deploy tasks > Run context](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/run-context.md) - Configuring a run with `flyte.with_runcontext()` - Execution target - Storage - Caching - Identity and resources - Logging - Code bundling - Context propagation - Reading context inside a task with `flyte.ctx()` - `TaskContext` fields - `ActionID` fields - Naming external resources - [Tasks > Run and deploy tasks > Entrypoint tasks](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/entrypoints.md) - When to use entrypoints - Mark a task as an entrypoint - Discover entrypoint tasks - CLI - Programmatic - UI - [Tasks > Run and deploy tasks > Run a Python script](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/run-python-script.md) - When to use it - Quickstart - What you can specify - Handling dependencies - 1. Standard library only - 2. Pip packages on top of the default image - 3. Bring your own image - Handling inputs and outputs - Passing inputs - Capturing outputs - Bundling: what gets uploaded - Python API - [Tasks > Run and deploy tasks > Run with notifications](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/run-with-notifications.md) - Execution phases - Template variables - Slack notifications - Email notifications - [Tasks > Run and deploy tasks > Rerun a run](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/rerun-runs.md) - Rerun from the UI - Rerun an entire run - Rerun a single action - Rerun from the CLI - Rerun compared with a fresh run - Rerun programmatically - Related - [Tasks > Run and deploy tasks > Recover a failed run](https://www.union.ai/docs/v2/flyte/user-guide/tasks/task-deployment/recover-runs.md) - What a recovery reuses - How this differs from caching - One verb, two behaviours - Recover from the CLI - Changing inputs - Recover programmatically - Related - [Apps](https://www.union.ai/docs/v2/flyte/user-guide/apps.md) - [Configure apps](configure-apps.md) - [Build apps](build-apps.md) - [Native app integrations](native-app-integrations.md) - [Serve and deploy apps](serve-and-deploy-apps.md) - Related - [Tasks](../tasks.md) - [Agents](../agents.md) - [Apps > Configure apps](https://www.union.ai/docs/v2/flyte/user-guide/apps/configure-apps.md) - Hello World example - Using fserve args - Using @app_env.server - Differences from TaskEnvironment - Configuration topics - [Apps > Configure apps > App environment settings](https://www.union.ai/docs/v2/flyte/user-guide/apps/configure-apps/app-environment-settings.md) - Shared environment settings - App-specific environment settings - Environment variable substitution in `args` - App startup - Server decorator via `@app_env.server` - Container command via `command` vs `args` - Shared settings - [Apps > Configure apps > Including additional files](https://www.union.ai/docs/v2/flyte/user-guide/apps/configure-apps/including-additional-files.md) - How include works - When to use include - Examples - Multi-file Streamlit app - Multi-file FastAPI app - App with configuration files - File discovery - Path resolution - Best practices - Limitations - [Apps > Configure apps > Passing parameters into app environments](https://www.union.ai/docs/v2/flyte/user-guide/apps/configure-apps/passing-parameters.md) - Parameter types overview - Basic parameter types - Accessing parameters in your app - `mount` - `env_var` - `get_parameter` - Full example - Delayed values - RunOutput - AppEndpoint - Overriding parameters at serve time - Example: FastAPI app with configurable model - Example: Using RunOutput for model serving - Best practices - Limitations - [Apps > Configure apps > /// script](https://www.union.ai/docs/v2/flyte/user-guide/apps/configure-apps/auto-scaling-apps.md) - Autoscaling apps - Scaling configuration - Basic scaling example - Scaling patterns - Idle TTL (Time to live) - Autoscaling best practices - Autoscaling limitations - Autoscaling troubleshooting - [Apps > Configure apps > Apps depending on other environments](https://www.union.ai/docs/v2/flyte/user-guide/apps/configure-apps/apps-depending-on-environments.md) - Basic usage - Example: App calling another app - Dependency chain - Multiple dependencies - Using AppEndpoint for dependency URLs - Deployment behavior - Task environment dependencies - Best practices - Example: A/B testing with dependencies - Limitations - [Apps > Build apps](https://www.union.ai/docs/v2/flyte/user-guide/apps/build-apps.md) - App types - Usage patterns - Next steps - [Apps > Build apps > Single-script apps](https://www.union.ai/docs/v2/flyte/user-guide/apps/build-apps/single-script-apps.md) - Plain Python HTTP server - Streamlit app - FastAPI app - Running single-script apps - When to use single-script apps - [Apps > Build apps > Multi-script apps](https://www.union.ai/docs/v2/flyte/user-guide/apps/build-apps/multi-script-apps.md) - FastAPI multi-script app - Project structure - Example: Multi-file FastAPI app - Automatic file discovery - Streamlit multi-script app - Project structure - Example: Multi-file Streamlit app - Deploying multi-file Streamlit app - Complex multi-file example - Project structure - Example code - Deploying complex app - Best practices - Troubleshooting - [Apps > Build apps > Serving graphs](https://www.union.ai/docs/v2/flyte/user-guide/apps/build-apps/serving-graphs.md) - Core concepts: a minimal two-app chain - Deploying multiple apps together with `depends_on` - Getting an upstream app's endpoint - Sizing each node independently - Example: CPU / GPU inference split - Disjoint images per node - GPU app: model.forward only - CPU app: pre/postprocess + call GPU - Deploy - Example: A/B testing with Statsig - Statsig client singleton - Variant apps - Root app with Statsig in its lifespan - App environments - Routing endpoint - Deploy - When to split into a serving graph - Best practices - [Apps > Build apps > Hybrid app-task graphs](https://www.union.ai/docs/v2/flyte/user-guide/apps/build-apps/hybrid-graphs.md) - Call app from task - Example: FastAPI app called from a task - Example: Call a model inference service from a task - Call task from app (webhooks / APIs) - Example: Basic webhook app - Advanced webhook patterns - Webhook security and best practices - Example: GitHub webhook - Gradio agent UI - Best practices - [Apps > Build apps > WebSocket apps](https://www.union.ai/docs/v2/flyte/user-guide/apps/build-apps/websocket-apps.md) - Example: Basic WebSocket app - WebSocket patterns - Using WebSockets with Flyte tasks - WebSocket client example - Best practices - [Apps > Build apps > Browser apps](https://www.union.ai/docs/v2/flyte/user-guide/apps/build-apps/browser-apps.md) - Accessing browser-based apps - Common browser-based app types - Streamlit apps - Gradio apps - Custom HTML/JS apps - Best practices - [Apps > Build apps > Secret-based authentication](https://www.union.ai/docs/v2/flyte/user-guide/apps/build-apps/secret-based-authentication.md) - Create the secret - Define the FastAPI app - Deploy the FastAPI app - Invoke the endpoint - Authentication for vLLM and SGLang apps - Create the authentication secret - Deploy vLLM app with authentication - Deploy SGLang app with authentication - Invoke authenticated LLM endpoints - Accessing Swagger documentation - Security best practices - Troubleshooting - Next steps - [Apps > Native app integrations](https://www.union.ai/docs/v2/flyte/user-guide/apps/native-app-integrations.md) - When to use a native integration - Available integrations - Next steps - [Apps > Native app integrations > Streamlit app](https://www.union.ai/docs/v2/flyte/user-guide/apps/native-app-integrations/streamlit-app.md) - Basic Streamlit app - Single-file Streamlit app - Multi-file Streamlit app - Example: Data visualization dashboard - Best practices - Troubleshooting - [Apps > Native app integrations > FastAPI app](https://www.union.ai/docs/v2/flyte/user-guide/apps/native-app-integrations/fastapi-app.md) - Basic FastAPI app - Serving a machine learning model - Accessing Swagger documentation - Example: REST API with multiple endpoints - Multi-file FastAPI app - Local-to-remote model serving - Best practices - Advanced features - Troubleshooting - [Apps > Native app integrations > vLLM app](https://www.union.ai/docs/v2/flyte/user-guide/apps/native-app-integrations/vllm-app.md) - Installation - Basic vLLM app - Using prefetched models - Model streaming - Custom vLLM arguments - Using the OpenAI-compatible API - Multi-GPU inference (Tensor parallelism) - Model sharding with prefetch - Autoscaling - Best practices - Troubleshooting - API reference - [Apps > Native app integrations > SGLang app](https://www.union.ai/docs/v2/flyte/user-guide/apps/native-app-integrations/sglang-app.md) - Installation - Basic SGLang app - Using prefetched models - Model streaming - Custom SGLang arguments - Using the OpenAI-compatible API - Multi-GPU inference (Tensor parallelism) - Model sharding with prefetch - Autoscaling - Structured generation - Best practices - Troubleshooting - API reference - [Apps > Native app integrations > Ollama app](https://www.union.ai/docs/v2/flyte/user-guide/apps/native-app-integrations/ollama-app.md) - Installation - Basic Ollama app - Choosing a model - Using the OpenAI-compatible API - Chat UI with Streamlit - Authentication - Autoscaling - Best practices - Troubleshooting - [Apps > Native app integrations > Flyte webhook](https://www.union.ai/docs/v2/flyte/user-guide/apps/native-app-integrations/flyte-webhook.md) - Available endpoints - Basic usage - Filtering endpoints - Endpoint groups - Individual endpoints - Allow-listing - Task allow-list - App allow-list - Trigger allow-list - Calling the webhook - Authentication - Self-reference protection - [Apps > Serve and deploy apps](https://www.union.ai/docs/v2/flyte/user-guide/apps/serve-and-deploy-apps.md) - Serve vs deploy - `flyte serve` - `flyte deploy` - Using Python SDK - Serve - Deploy - Using the CLI - Serve - Deploy - Next steps - [Apps > Serve and deploy apps > How app serving works](https://www.union.ai/docs/v2/flyte/user-guide/apps/serve-and-deploy-apps/how-app-serving-works.md) - Overview - Using the Python SDK - Overriding parameters - Advanced serving options - Using CLI - Return value - Best practices - Troubleshooting - [Apps > Serve and deploy apps > How app custom domains work](https://www.union.ai/docs/v2/flyte/user-guide/apps/serve-and-deploy-apps/how-app-custom-domain-works.md) - [Apps > Serve and deploy apps > How app deployment works](https://www.union.ai/docs/v2/flyte/user-guide/apps/serve-and-deploy-apps/how-app-deployment-works.md) - Overview - Using the Python SDK - Deployment plan - Overriding app configuration at deployment time - Activation/deactivation - Using the CLI - Example: Full deployment configuration - Best practices - Deployment status and return value - Troubleshooting - [Apps > Serve and deploy apps > Activating and deactivating apps](https://www.union.ai/docs/v2/flyte/user-guide/apps/serve-and-deploy-apps/activating-and-deactivating-apps.md) - Activation - Activate after deployment - Activate an app - Check activation status - Deactivation - Lifecycle management - Typical deployment workflow - Blue-green deployment - Using CLI - Activate - Deactivate - Check status - Best practices - Automatic activation with serve - Example: Complete deployment and activation - Troubleshooting - [Apps > Serve and deploy apps > Prefetching models](https://www.union.ai/docs/v2/flyte/user-guide/apps/serve-and-deploy-apps/prefetching-models.md) - Why prefetch? - Basic prefetch - Using Python SDK - Using CLI - Using prefetched models - Prefetch options - Custom artifact name - With HuggingFace token - With resources - Sharding models for multi-GPU - vLLM sharding - Using shard config via CLI - Using prefetched sharded models - CLI options - Complete example - Best practices - Troubleshooting - [Triggers](https://www.union.ai/docs/v2/flyte/user-guide/triggers.md) - Types of automation - In this section - [Configure a trigger](configure-triggers.md) - [Schedule triggers](schedules.md) - [When scheduled runs start](schedule-timing.md) - [Manual triggers and firing on demand](manual-triggers.md) - [Notifications](trigger-notifications.md) - [Deploy and manage triggers](manage-triggers.md) - [Triggers > Configure a trigger](https://www.union.ai/docs/v2/flyte/user-guide/triggers/configure-triggers.md) - Triggers are set in the task decorator - `flyte.Trigger` - The `inputs` parameter - Basic usage - Using `flyte.TriggerTime` - Required vs optional parameters - Complex input types - Multiple triggers per task - [Triggers > Schedule triggers](https://www.union.ai/docs/v2/flyte/user-guide/triggers/schedules.md) - Fixed-rate schedules with `flyte.FixedRate` - Examples - Cron schedules with `flyte.Cron` - Examples - Predefined schedule triggers - Available predefined triggers - Trigger time in predefined triggers - Schedule time zones - Setting time zone for a cron schedule - `flyte.TriggerTime` is always in UTC - Daylight savings time behavior - [Triggers > When scheduled runs start](https://www.union.ai/docs/v2/flyte/user-guide/triggers/schedule-timing.md) - Cron-based triggers - Fixed-rate triggers without `start_time` - No `start_time`, auto_activate: True - No `start_time`, auto_activate: False - Fixed-rate triggers with `start_time` - Fixed-rate with `start_time` while active - Fixed-rate with `start_time` while inactive - [Triggers > Manual triggers and firing on demand](https://www.union.ai/docs/v2/flyte/user-guide/triggers/manual-triggers.md) - Triggers without automation - Programmatic - CLI - Programmatic - CLI - Firing a trigger on demand - From the UI - From Python - [Triggers > Notifications](https://www.union.ai/docs/v2/flyte/user-guide/triggers/trigger-notifications.md) - Execution phases - Template variables - Slack notifications - Email notifications - Microsoft Teams notifications - Custom webhook notifications - [Triggers > Deploy and manage triggers](https://www.union.ai/docs/v2/flyte/user-guide/triggers/manage-triggers.md) - Deploying a task with triggers - Programmatic - CLI - Activating and deactivating triggers - Programmatic - CLI - Deleting triggers - Programmatic - CLI - [Agents](https://www.union.ai/docs/v2/flyte/user-guide/agents.md) - [Build an agent](build-agent.md) - [Build an MCP](build-mcp.md) - [Sandboxing](sandboxing.md) - Related - [Agent frameworks](../../integrations/agents.md) - [Tasks](../tasks.md) - [System one types](../system-one-types.md) - [Agents > Build an agent](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-agent.md) - How Flyte maps to the agentic world - Ways to build an agent - Deploying an agent - Related - [Agents > Build an agent > Pure Python agents](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-agent/python-agents.md) - ReAct pattern: Reason, Act, Observe (no framework needed) - Plan-and-Execute with parallel fan-out - More agentic patterns - How Flyte's primitives map to the agent stack - Next steps - [Agents > Build an agent > Flyte-native agents](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-agent/flyte-agents.md) - How it works - Sync vs async - A minimal agent - Tools - Customizing a tool with `tool(...)` - MCP integration - Skills - Observability - Extending the agent class - `run` is sync-by-default - Strategy 1: wrap the built-in loop - Strategy 2: implement `run` from scratch - Choosing between subclassing and composition - Next steps - [Agents > Build an agent > Bring your own framework](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-agent/bring-your-own-framework.md) - How much control does the framework give you? - The core pattern - Make tools durable - Trace the framework's internals - Fan out across containers - Checklist - Next steps - [Agents > Build an agent > Agent memory](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-agent/agent-memory.md) - What a `MemoryStore` holds - Sync vs async - Keyed stores: the easy path - Working with a MemoryStore independently - Optimistic concurrency - Optional capabilities - Passing memory to the agent - Lower-level usage - Next steps - [Agents > Build an agent > Agent chat UI](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-agent/agent-chat-ui.md) - Option 1: the built-in chat UI - Option 2: a custom FastAPI chat app - Next steps - [Agents > Build an agent > Deploy an agent as a service](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-agent/deploy-agent-as-service.md) - As a task - As a scheduled task (via `Trigger`) - Behind a webhook (`AppEnvironment`) - Chat and other app patterns - [Agents > Build an MCP](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-mcp.md) - HTTP layout - Quickstart - [Agents > Build an MCP > User-defined MCP server](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-mcp/mcp_server.md) - When to use it - How it works - Basic example - HTTP layout - Choosing a transport - Connecting a client - Claude Code - OpenCode - Configuration tips - Best practices - [Agents > Build an MCP > Flyte MCP server](https://www.union.ai/docs/v2/flyte/user-guide/agents/build-mcp/flyte_mcp_server.md) - When to use it - How to run it - Running locally with `uvx` - Basic example - Scoping the server - 1. Tool groups - 2. Individual tools - 3. Allowlists - Enabling the search tools - Putting it together: a filtered server - Connecting a client - Claude Code: local (stdio) - OpenCode: local - Best practices - MCP tools reference - [Agents > Sandboxing](https://www.union.ai/docs/v2/flyte/user-guide/agents/sandboxing.md) - Why sandboxing matters for AI - Types of sandboxes - What Flyte offers - Workflow sandbox (Monty) - Code sandbox (container) - When to use which - Learn more - [Agents > Sandboxing > Workflow sandboxing in Flyte](https://www.union.ai/docs/v2/flyte/user-guide/agents/sandboxing/workflow-sandboxing-flyte.md) - Why workflow sandboxing? - How it works - Example: sandboxed orchestrator - Example: dynamic code execution - Reusable task from a code string - One-shot local execution - Parameterized code generation - Building agents with programmatic tool calling - Syntax restrictions - Allowed - Not allowed - Type restrictions - Security model - [Agents > Sandboxing > Programmatic tool calling for agents](https://www.union.ai/docs/v2/flyte/user-guide/agents/sandboxing/code-mode.md) - Programmatic tool calling vs sequential tool calling - Why programmatic tool calling is powerful - Token efficiency - Performance - Natural programming patterns - Progressive tool discovery - Data privacy - Example: sequential vs programmatic tool calling - Sequential tool calling approach - Programmatic tool calling approach - Example: defining tools - Example: programmatic tool-calling agent - Example: chat app - Example: durable agent - References - [Agents > Sandboxing > Code sandboxing](https://www.union.ai/docs/v2/flyte/user-guide/agents/sandboxing/code-sandboxing.md) - Execution modes - Auto-IO mode - Verbatim mode - Command mode - Executing a sandbox - Error handling - Supported types - How types are handled - Configuring the container image - Python packages - System packages - Additional Dockerfile commands - Pre-built images - Image configuration - Runtime configuration - Resources - Retries - Timeout - Environment variables - Secrets - Caching - Deploying a sandbox as a task - End-to-end example - API reference - `flyte.sandbox.create()` - Sandbox methods - [System one types](https://www.union.ai/docs/v2/flyte/user-guide/system-one-types.md) - The types - The battery - Two models, two jobs - Where it fits - In a data pipeline - In an agent loop - In an app or service - The discipline that makes it work - When not to reach for one - Implementations - [Typed decisions with Jev](typed-decisions-with-jev.md) - Worked example - [Typed decisions for agentic pipelines](../../tutorials/agents/system-one-agents.md) - Related - [System one types > Typed decisions with Jev](https://www.union.ai/docs/v2/flyte/user-guide/system-one-types/typed-decisions-with-jev.md) - The three field types - Declare the target type - Coerce - Derive the verdict from the fields - Gate on confidence, and mean it - Coerce the agent's next move - Coerced values cross task boundaries - Related - [Project patterns](https://www.union.ai/docs/v2/flyte/user-guide/project-patterns.md) - [Bring your own image (BYOI)](bring-your-own-image.md) - [Monorepo with uv](monorepo-with-uv.md) - [CI/CD deployments](cicd.md) - [Project patterns > Bring your own image](https://www.union.ai/docs/v2/flyte/user-guide/project-patterns/bring-your-own-image.md) - The multi-team problem - Pattern 1: Pure BYOI - Dockerfiles - Build and push - Python code - Run and deploy - Pattern 2: Remote builder - The base images - Adapting with `flyte.Image` - Task definitions - Entry point - Run and deploy - Base image USER and WORKDIR requirements - Decision matrix - [Project patterns > Structuring Flyte projects with uv](https://www.union.ai/docs/v2/flyte/user-guide/project-patterns/monorepo-with-uv.md) - The two layers - How the image gets built - `with_code_bundle()`: one image for dev and prod - `root_dir` - Monorepo patterns - Pattern A: Shared lockfile (recommended) - Pattern B: Independent packages - Pattern C: uv workspace (`[tool.uv.workspace]`) - The full build path (production) - [Project patterns > CI/CD deployments](https://www.union.ai/docs/v2/flyte/user-guide/project-patterns/cicd.md) - What CI needs to do - Authentication: client credentials - Register a client-credentials application - Store the secret as a CI secret - Point the CLI at the credential - Scope and rotation - Project configuration - `config.yaml` - The GitHub Actions workflow - Key flag choices - Splitting build from deploy - [Software development lifecycle](https://www.union.ai/docs/v2/flyte/user-guide/software-development-lifecycle.md) - Why data and AI systems need more than CI - What Flyte provides - Integrations by stage - Scope of this section - [Software development lifecycle > Event-driven automation](https://www.union.ai/docs/v2/flyte/user-guide/software-development-lifecycle/event-driven-automation.md) - Choose the cause - Receive a webhook - Launch once per event - Keep the receiver thin - Restrict scopes - Test in stages - See also - [Software development lifecycle > Review and release gates](https://www.union.ai/docs/v2/flyte/user-guide/software-development-lifecycle/review-and-release-gates.md) - Match the gate to the change - Gate a pull request - Gate on a person's review - Promote the same version - Gate agent pull requests - Gate a dataset change - Gates to avoid - See also - [Software development lifecycle > Human gates and approvals](https://www.union.ai/docs/v2/flyte/user-guide/software-development-lifecycle/human-gates-and-approvals.md) - Choose where the person answers - Ask in Slack - Set a timeout - Make the prompt decidable - Control who can answer - Keep human gates rare - Review AI output by exception - See also - [Software development lifecycle > Evaluation gates](https://www.union.ai/docs/v2/flyte/user-guide/software-development-lifecycle/evaluation-gates.md) - Build the gate - Why a test suite doesn't fit - Compare honestly - Record the result - Gate the input - Scale the compute - Record the configuration - Evaluate agents - See also - [Software development lifecycle > Notifications and observability](https://www.union.ai/docs/v2/flyte/user-guide/software-development-lifecycle/notifications-and-observability.md) - Notify on a terminal state - Post from inside a run - Reply to the person who asked - Observe across runs - Track what produced an output - Investigate a failure - File the result in your tracker - See also - [Scale your runs](https://www.union.ai/docs/v2/flyte/user-guide/run-scaling.md) - Understanding Flyte execution - Performance optimization - Key concepts for scaling - [Scale your runs > Data flow](https://www.union.ai/docs/v2/flyte/user-guide/run-scaling/data-flow.md) - Overview - Data types and transport - Passed by reference - Passed by value (inline I/O) - Task execution and data flow - Input download - Output upload - Task-to-task data flow - Caching and data hashing - Cache key computation - Inline data caching - Raw data hashing - Cache control - Traces and data flow - Object stores and latency considerations - Configuring data storage - Organization and project level - Per-run configuration - [Scale your runs > Life of a run](https://www.union.ai/docs/v2/flyte/user-guide/run-scaling/life-of-a-run.md) - Overview - Phase 1: Code analysis and preparation - Phase 2: Image building - Phase 3: Code bundling - Default: `copy_style="loaded_modules"` - Alternative: `copy_style="none"` - Phase 4: Upload code bundle - Phase 5: Run creation and queuing - Phase 6: Task execution in data plane - Container startup - Invoking downstream tasks - Execution flow diagram - Action identifiers and crash recovery - Downstream task execution - Reusable containers - Reusable container execution flow - State replication and visualization - Queue service to run service - UI limitations - Optimization opportunities - [Scale your runs > Scale your workflows](https://www.union.ai/docs/v2/flyte/user-guide/run-scaling/scale-your-workflows.md) - Understanding performance dimensions - Latency - Throughput - Task execution overhead - The overhead principle - System architecture and data flow - Optimization strategies - 1. Use reusable containers for concurrency - 2. Batch workloads to reduce overhead - 3. Use traces for lightweight operations - 4. Limit fanout for system stability - 5. Optimize data transfer - 6. Use caching - 7. Parallelize with `flyte.map` - Performance tuning workflow - Real-world example: PyIceberg batch processing - Example: Optimizing a data pipeline - Before optimization - After optimization - When to contact the Union team - [Scale your runs > Maximize GPU utilization for batch inference](https://www.union.ai/docs/v2/flyte/user-guide/run-scaling/batch-inference.md) - Why GPU utilization drops - Serving vs in-process batch inference - Solution: `DynamicBatcher` - Basic usage - Cost estimation - `TokenBatcher` for LLM inference - Combining with app environments - Example: batch LLM inference with vLLM behind a FastAPI app - Monitoring utilization - [Advanced project: LLM reporting agent](https://www.union.ai/docs/v2/flyte/user-guide/advanced-project.md) - What you'll build - Concepts covered - Architecture - Prerequisites - Parts - Key takeaways - [Advanced project: LLM reporting agent > Resilient generation](https://www.union.ai/docs/v2/flyte/user-guide/advanced-project/resilient-generation.md) - Two environments - Reusable environment for LLM work - ReusePolicy parameters - Standard environment for orchestration - Traced LLM calls - Benefits of tracing - When to use @flyte.trace - Traced helper functions - Retry strategies - Configuring retries - Combining tracing with retries - Structured prompts - Pydantic models for structured output - Next steps - [Advanced project: LLM reporting agent > Agentic refinement](https://www.union.ai/docs/v2/flyte/user-guide/advanced-project/agentic-refinement.md) - The agentic pattern - Critique function - Revise function - The refinement loop - How it works - Early exit - Grouping iterations with flyte.group - Why use flyte.group? - Group context - Configuring the loop - Choosing thresholds - Best practices for agentic loops - Next steps - [Advanced project: LLM reporting agent > Parallel outputs](https://www.union.ai/docs/v2/flyte/user-guide/advanced-project/parallel-outputs.md) - The formatting functions - When to trace and when not to - Parallel execution with asyncio.gather - How asyncio.gather works - When to use asyncio.gather - Grouping parallel operations - Collecting outputs in a directory - The batch pipeline - Pipeline flow - Running the pipeline - Cost optimization tips - 1. Choose the right model - 2. Tune iteration parameters - 3. Use caching effectively - 4. Scale the batch - Next steps - [Advanced project: LLM reporting agent > Serving app](https://www.union.ai/docs/v2/flyte/user-guide/advanced-project/serving-app.md) - App environment configuration - Key configuration - Connecting to pipeline output with RunOutput - The Streamlit application - Displaying multiple reports - Generation instructions - Deploying the app - Workflow: Generate then view - Automatic updates with RunOutput - Complete example structure - Running the complete example - Summary - [Migration](https://www.union.ai/docs/v2/flyte/user-guide/migration.md) - [From Flyte 1 to 2](flyte-2.md) - [From Airflow to Flyte](from-airflow.md) - [From Slurm to Flyte](from-slurm.md) - [Migration > From Flyte 1 to 2](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2.md) - Pure Python execution - Sync Python - Async Python - Simplified API - Fine-grained reproducibility and recoverability - Improved remote functionality - Native notebook support - [Migration > From Flyte 1 to 2 > Migration overview](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/overview.md) - Pure Python execution - Flyte 1 - Flyte 2 - Asynchronous model - True parallelism for all workloads - Calling sync tasks from async tasks - The `flyte.map` function: Familiar patterns - Sync Map - Async Map - Terminology and concept mapping - Package imports - Flyte 1 - Flyte 2 - The two mechanical changes behind (almost) every migration - 1. Move task configuration into a `TaskEnvironment` - 2. Replace `@task` / `@workflow` / `@dynamic` with `@env.task` - In this section - Quick reference - [Migration > From Flyte 1 to 2 > Tasks and workflows](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/tasks-and-workflows.md) - Hello world: tasks and workflows - Flyte 1 - Flyte 2 - Chaining and ordering - Flyte 1 - Flyte 2 - Subworkflows - Flyte 1 - Flyte 2 - TaskEnvironment configuration - Parameter mapping: `@task` → `TaskEnvironment` + `@env.task` - Next - [Migration > From Flyte 1 to 2 > Task configuration](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/configuration.md) - Image, resources, and caching - Flyte 1 - Flyte 2 - Container images - Resources - Caching - Secrets - Flyte 1 - Flyte 2 - Scheduling - Flyte 1 - Flyte 2 - Next - [Migration > From Flyte 1 to 2 > CLI and configuration](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/cli-and-configuration.md) - CLI command mapping - Running tasks - Flyte 1 - Flyte 2 - Deploying - Flyte 1 - Flyte 2 - Key flag differences - Configuration files - Flyte 1 - Flyte 2 - Configuring in code - Next - [Migration > From Flyte 1 to 2 > Control flow](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/control-flow.md) - Conditional execution - Flyte 1 - Flyte 2 - Dynamic workflows - Flyte 1 - Flyte 2 - Error handling - Flyte 1 - Flyte 2 - Next - [Migration > From Flyte 1 to 2 > Parallelism and fan-out](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/parallelism.md) - Fan-out: `map_task` - Flyte 1 - Flyte 2 (flyte.map) - Flyte 2 (asyncio.gather) - Choosing `flyte.map` vs `asyncio.gather` - Concurrency control and error handling - Data backfills - Flyte 1 - Flyte 2 - Next - [Migration > From Flyte 1 to 2 > Data types and I/O](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/data-io.md) - Files and directories - Flyte 1 - Flyte 2 - DataFrames - Flyte 1 - Flyte 2 - Dataclasses and structured types - Flyte 1 - Flyte 2 - Data ETL - Flyte 1 - Flyte 2 - Next - [Migration > From Flyte 1 to 2 > ML workloads](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/ml-workloads.md) - Small model training (scikit-learn / XGBoost) - Flyte 1 - Flyte 2 - Hyperparameter optimization - Flyte 1 - Flyte 2 - Large model training (deep learning) - Flyte 1 - Flyte 2 - Batch inference - Flyte 1 - Flyte 2 - A complete example: end-to-end ML pipeline - Flyte 1 - Flyte 2 - Next - [Migration > From Flyte 1 to 2 > New in Flyte 2: patterns that weren't possible in Flyte 1](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/new-in-flyte-2.md) - Real-time inference and model serving - LLM serving - App serving (dashboards and APIs) - Dynamic batching for GPU inference - Sandboxed code execution - Next - [Migration > From Flyte 1 to 2 > Hybrid v1 and v2 pipelines](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/hybrid-pipelines.md) - Running a Flyte 2 task from a Flyte 1 workflow - 1. Create the API key and store it - 2. Write the bridging task - Running a Flyte 1 workflow from a Flyte 2 task - Considerations - See also - [Migration > From Flyte 1 to 2 > Gotchas and caveats](https://www.union.ai/docs/v2/flyte/user-guide/migration/flyte-2/gotchas-and-caveats.md) - Common gotchas - Non-deterministic behavior - Dealing with non-determinism - Type safety - No global state - Driver pod requirements - OOM risk from materialized I/O - [Migration > From Airflow to Flyte](https://www.union.ai/docs/v2/flyte/user-guide/migration/from-airflow.md) - [Part 1: Vanilla Operators](part-1-vanilla-operators.md) - [Migration > From Airflow to Flyte > Part 1: vanilla operators](https://www.union.ai/docs/v2/flyte/user-guide/migration/from-airflow/part-1-vanilla-operators.md) - 1. Where dependencies are specified - 2. The driver task (in place of DAGs) - 3. Triggers (in place of schedules) - 4. PythonOperator to `@env.task` - File and Dir: for data that doesn't fit in a return value - 5. TaskFlow to `@env.task` - TaskFlow decorator variants - 6. BashOperator to ContainerTask - When to use `ContainerTask` - How the arguments map - 7. KubernetesPodOperator to TaskEnvironment + PodTemplate - Where every KPO knob lands - What a fully-specified task looks like - 8. Orchestration: parallelism, conditionals, error handling - Parallelism - Dynamic mapping - Conditionals - Error handling - What's next - Caching - Reusable containers - Reports - Apps - [Migration > From Slurm to Flyte](https://www.union.ai/docs/v2/flyte/user-guide/migration/from-slurm.md) - The mapping at a glance - The job script becomes a task - `module load` becomes an image - Job arrays become fan-out - Job dependencies become ordinary Python - `--requeue` becomes retries, spot handling, and checkpoints - Multi-node jobs become clustered tasks - The shared filesystem becomes explicit data - Watching and debugging jobs - Staging the migration - Scheduler features without an equivalent - [Telemetry](https://www.union.ai/docs/v2/flyte/user-guide/telemetry.md) - What is collected - Turn off telemetry on your machine - Turn off telemetry in tasks and apps - Turn off telemetry in a deployment - Variables reference --- ## Tutorials - [Biotech & healthcare](https://www.union.ai/docs/v2/flyte/tutorials/biotech-healthcare.md) - [Genomic alignment](genomic-alignment.md) - [Cross-species gene comparison](genomic-gene-comparison.md) - [Genomic variant effect prediction](genomic-variant-effect.md) - [Brain tumor MRI classification](tumor-detection.md) - [Drug molecule screening agent](drug-molecule-screening.md) - [Biotech & healthcare > Genomic alignment](https://www.union.ai/docs/v2/flyte/tutorials/biotech-healthcare/genomic-alignment.md) - Define the container image - Define the task environments - Define the data classes - Fetch assets - Quality filtering with fastp - Build the Bowtie 2 index - Align reads - Orchestrate the workflow - Run the workflow - [Biotech & healthcare > Brain tumor MRI classification](https://www.union.ai/docs/v2/flyte/tutorials/biotech-healthcare/tumor-detection.md) - Define the container image - Define the task environments - Configure training - Load the dataset - Train the model - Resumable checkpointing - Generate the report - Orchestrate the pipeline - Run the pipeline - [Biotech & healthcare > Cross-species gene comparison](https://www.union.ai/docs/v2/flyte/tutorials/biotech-healthcare/genomic-gene-comparison.md) - Define the task environments - Orchestrate the pipeline - Run the workflow - [Biotech & healthcare > Genomic variant effect prediction](https://www.union.ai/docs/v2/flyte/tutorials/biotech-healthcare/genomic-variant-effect.md) - Define the task environments - Orchestrate the pipeline - Run the workflow - [Biotech & healthcare > Drug molecule screening agent](https://www.union.ai/docs/v2/flyte/tutorials/biotech-healthcare/drug-molecule-screening.md) - Define the task environment - Define the screening agent - Run the agentic pipeline - Two-round rescreen demo (complex execution graph) - [Geospatial](https://www.union.ai/docs/v2/flyte/tutorials/geospatial.md) - [GPU-accelerated climate modeling](climate-modeling.md) - [Geospatial > GPU-accelerated climate modeling](https://www.union.ai/docs/v2/flyte/tutorials/geospatial/climate-modeling.md) - Overview - Implementation - Dependencies and container image - Simulation parameters and data structures - Task environments - Data ingestion: multiple sources in parallel - Preprocessing with Dask - GPU-accelerated atmospheric simulation - Distributing across multiple GPUs - The main workflow - Running the pipeline - Key concepts - Ensemble forecasting - Adaptive mesh refinement - Real-time event detection - Where to go next - [Financial services & fintech](https://www.union.ai/docs/v2/flyte/tutorials/financial-services.md) - [Financial research agent](financial-research-agent.md) - [Fraud detection with Feast](fraud-detection-feast.md) - [Multi-agent trading simulation](trading-agents.md) - [Financial services & fintech > Multi-agent trading simulation](https://www.union.ai/docs/v2/flyte/tutorials/financial-services/trading-agents.md) - TL;DR - What is an agent, anyway? - What's different here? - How it works: step-by-step walkthrough - Entry point - Analyst agents - Research agents - Trading agent - Risk agents - Retaining agent memory with S3 vectors - Running the simulation - Why Flyte? _(A quick note before you go)_ - [Financial services & fintech > Fraud detection with Feast](https://www.union.ai/docs/v2/flyte/tutorials/financial-services/fraud-detection-feast.md) - Define the task environment - Orchestrate the pipeline - Run the workflow - [Financial services & fintech > Financial research agent](https://www.union.ai/docs/v2/flyte/tutorials/financial-services/financial-research-agent.md) - Setting up the environment - Data types - You.com Research and Search APIs - Synthesize briefings with Claude - Research one company - Orchestration - Run the agent - Create secrets - Run locally or remotely - [Frontier AI](https://www.union.ai/docs/v2/flyte/tutorials/frontier-ai.md) - [Distributed LLM pretraining](distributed-pretraining.md) - [Durable agentic RL with SkyRL and Harbor](agentic-rl-skyrl.md) - [Frontier AI > Distributed LLM pretraining](https://www.union.ai/docs/v2/flyte/tutorials/frontier-ai/distributed-pretraining.md) - Overview - Implementation - Setting up the environment - Declaring resource requirements - Model configurations - Building the GPT model - The Lightning training module - Checkpointing for fault tolerance - Real-time metrics with Flyte reports - Streaming data at scale - Distributed training with FSDP - Tying it together - Running the pipeline - Going further - [Frontier AI > Durable agentic RL with SkyRL and Harbor](https://www.union.ai/docs/v2/flyte/tutorials/frontier-ai/agentic-rl-skyrl.md) - Overview - Implementation - Environments: one failure domain per tier - Weights travel as inputs (pull-based sync) - Worlds: built once, private copy per trial - The judge is memoized: retried, never re-billed - Training steps that can't double-apply - The driver: a durable log of the training run - Re-scoring old rollouts with a new rubric - Run it - What this means for real training - [Computer vision](https://www.union.ai/docs/v2/flyte/tutorials/computer-vision.md) - [Fine-tuning a VLM](qwen-vl-finetuning.md) - [RT-DETR object detection](detr-object-detection.md) - [Multimodal retrieval evaluation](multimodal-retrieval-evaluation.md) - [Computer vision > Fine-tuning a VLM](https://www.union.ai/docs/v2/flyte/tutorials/computer-vision/qwen-vl-finetuning.md) - Overview - Implementation - Setting up the environment - Preparing the dataset - The adapter - Multi-node training with DeepSpeed - Fault tolerance and recovery - Live observability - Evaluation - Putting it all together - Running the tutorial - Going further - [Computer vision > Multimodal retrieval evaluation](https://www.union.ai/docs/v2/flyte/tutorials/computer-vision/multimodal-retrieval-evaluation.md) - Define the container image - Define the task environments - Configuration and data types - Loading, indexing, and search - Run one experiment - Compare experiments - Run the evaluation - [Computer vision > RT-DETR object detection](https://www.union.ai/docs/v2/flyte/tutorials/computer-vision/detr-object-detection.md) - Define the task environments - Orchestrate the pipeline - Run the workflow - [Agents](https://www.union.ai/docs/v2/flyte/tutorials/agents.md) - [Autoresearch agent](autoresearch.md) - [Parallelized autoresearch agent](parallelized-autoresearch-agent.md) - [Code mode analytics agent](code-mode-agent.md) - [AutoSec researcher agent](autosec-research-agent.md) - [Coding agent](code-agent.md) - [Competitive intelligence agent](competitive-intelligence-agent.md) - [Compliance monitoring agent](compliance-monitoring-agent.md) - [Deep research](deep-research.md) - [LangGraph research agent](langgraph-agent-research.md) - [Field data enrichment agent](field-data-enrichment-agent.md) - [MLE Bot: autonomous ML engineer](mle-bot.md) - [Support resolution agent](support-resolution-agent.md) - [Typed decisions for agentic pipelines](system-one-agents.md) - [Agents > Autoresearch agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/autoresearch.md) - Define the container image - Define the task environment - What changed - All changed files - Model the result - What changed - All changed files - The autoresearch task - What changed - All changed files - What changed - All changed files - Run the agent - Create secrets - Prepare the research repository - Run remotely - [Agents > Parallelized autoresearch agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/parallelized-autoresearch-agent.md) - Define the task environments - Right-size experiments with `call_handler` - The fan-out agent task - Run the agent - Create secrets - Run remotely - [Agents > AutoSec researcher agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/autosec-research-agent.md) - Define the task environment - Run the security pipeline - Run the agent - Create secrets - Run remotely - [Agents > Coding agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/code-agent.md) - What this example demonstrates - Setting up the agent environment - Retrieve docs - Code generation - Running the code agent - [Agents > Competitive intelligence agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/competitive-intelligence-agent.md) - Setting up the environment - Data types - Search with the You.com Search API - Extract deltas with Claude - Watch one competitor - Orchestration - Run the agent - Create secrets - Run locally or remotely - [Agents > Deep research](https://www.union.ai/docs/v2/flyte/tutorials/agents/deep-research.md) - Setting up the environment - Generate research queries - Search and summarize - Evaluate research completeness - Filter results - Generate the final answer - Orchestration - Run the deep research agent - Evaluate with Weights & Biases Weave - [Agents > LangGraph research agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/langgraph-agent-research.md) - Define the task environment - Orchestrate the pipeline - Run the agent - [Agents > MLE bot: an autonomous ML engineer](https://www.union.ai/docs/v2/flyte/tutorials/agents/mle-bot.md) - TL;DR - The problem with LLMs and ML pipelines - How it works - What to expect - Declaring task environments - Building durable tool functions - Guiding the LLM with domain knowledge - Dataset context - General ML best practices - The agent loop: profile, design, execute, iterate - Dataset context - General ML best practices — apply these based on the dataset context above - Available tools - Monty sandbox restrictions - Critical patterns for using tool results - When fixing a previous error - Pipeline design — you decide the structure - Response format - Reasoning - Code - Running LLM-generated code in Flyte's sandbox - Dataset context - General ML best practices — apply these based on the dataset context above - Available tools - Monty sandbox restrictions - Critical patterns for using tool results - When fixing a previous error - Pipeline design — you decide the structure - Response format - Reasoning - Code - Streaming results to a live report - Dataset context - General ML best practices — apply these based on the dataset context above - Available tools - Monty sandbox restrictions - Critical patterns for using tool results - When fixing a previous error - Pipeline design — you decide the structure - Response format - Reasoning - Code - Running it - Why Flyte? - [Agents > Compliance monitoring agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/compliance-monitoring-agent.md) - Setting up the environment - Data types - Research with the You.com Research API - Triage findings with Claude - Monitor one watch item - Orchestration - Run the agent - Create secrets - Run locally or remotely - [Agents > Field data enrichment agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/field-data-enrichment-agent.md) - Setting up the environment - Data types - Search with the You.com Search API - Enrich one event - Orchestration - Run the agent - Create secrets - Run locally or remotely - [Agents > Support resolution agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/support-resolution-agent.md) - Setting up the environment - Data types - Ground answers with the You.com Research API - Ground one ticket - Draft a customer-ready reply - Resolve one ticket - Orchestration - Run the agent - Create secrets - Run locally or remotely - [Agents > Code mode analytics agent](https://www.union.ai/docs/v2/flyte/tutorials/agents/code-mode-agent.md) - Why code mode - What runs where - The heavy tool: a durable query - Tools are a safety boundary - The durable environment - The native agent - The report collector - Live prices over MCP - The analysis task - Serving it: the native chat app - Deploy and run - Going further - [Agents > Typed decisions for agentic pipelines](https://www.union.ai/docs/v2/flyte/tutorials/agents/system-one-agents.md) - The four patterns it is built around - Setting up the environment - The vocabulary - The battery - Composing the verdict - The tools, and the cases - Pattern 1: System One as a typed guard - Pattern 2: plan, fan out, aggregate - Pattern 3: a loop whose control flow is typed - The benchmark - What to look at in the report - Why the baseline has to owe the whole battery - Secrets - Adapting it to your own task - Related - [Context engineering](https://www.union.ai/docs/v2/flyte/tutorials/context-engineering.md) - [Automatic prompt engineering](auto_prompt_engineering.md) - [Text-to-SQL](text_to_sql.md) - [Context engineering > Text-to-SQL prompt optimization](https://www.union.ai/docs/v2/flyte/tutorials/context-engineering/text_to_sql.md) - Ingesting data - From question to SQL - Vector indexing - Table retrieval and context building - SQL generation and response synthesis - Building the QA dataset - Schema extraction and chunking - Question and SQL generation - Validation and quality control - Optimizing prompts - Evaluation pipeline - Iterative optimization - Run it - What we observed - The bigger lesson - [Context engineering > Automatic prompt engineering](https://www.union.ai/docs/v2/flyte/tutorials/context-engineering/auto_prompt_engineering.md) - Set up the environment - Prepare the evaluation dataset - Define models - Evaluate prompts - Optimize prompts - Build the full pipeline - Run it - Why this matters - Next steps - [Model training](https://www.union.ai/docs/v2/flyte/tutorials/model-training.md) - [Hyperparameter optimization](hpo.md) - [LLM fine-tuning with LoRA and QLoRA](llm-fine-tuning-lora-qlora.md) - [BERT emotion classification](bert-fine-tuning-emotion.md) - [Model training > LLM fine-tuning with LoRA and QLoRA](https://www.union.ai/docs/v2/flyte/tutorials/model-training/llm-fine-tuning-lora-qlora.md) - Define the task environments - Orchestrate the pipeline - Run the workflow - [Model training > BERT emotion classification](https://www.union.ai/docs/v2/flyte/tutorials/model-training/bert-fine-tuning-emotion.md) - Define the task environments - Orchestrate the pipeline - Run the workflow - [Model training > Hyperparameter optimization](https://www.union.ai/docs/v2/flyte/tutorials/model-training/hpo.md) - A better way to run HPO - Declare dependencies - Define the task environment - Define the optimizer - Define the objective function - Define the main optimization loop - Run the experiment --- ## Integrations - [Agent frameworks](https://www.union.ai/docs/v2/flyte/integrations/agents.md) - Supported frameworks - Two decorators - Quick start - What Flyte adds - Capability matrix - Choosing a framework - Next steps - [Agent frameworks > How it works](https://www.union.ai/docs/v2/flyte/integrations/agents/how-it-works.md) - The division of labor - Tools are Flyte tasks - Passing tools - Renaming a tool - Durable model turns - What replay means in practice - Where durability does not reach - Cross-run memory - Observability - Human in the loop - Multi-agent orchestration - Sync and async - Running locally - The shared contract - Next steps - [Agent frameworks > OpenAI Agents SDK](https://www.union.ai/docs/v2/flyte/integrations/agents/openai.md) - Installation - Quick start - How it maps to Flyte - Bring your own agent - Custom run configuration - Memory - Building blocks - `run_agent` parameters - Notes - Examples - [Agent frameworks > Claude Agent SDK](https://www.union.ai/docs/v2/flyte/integrations/agents/claude-agent-sdk.md) - Installation - Quick start - How it maps to Flyte - Durability - Observability - Bring your own options - Human in the loop - Memory - `run_agent` parameters - Examples - [Agent frameworks > Google ADK](https://www.union.ai/docs/v2/flyte/integrations/agents/google-adk.md) - Installation - Quick start - How it maps to Flyte - Bring your own agent - Memory - Bounding a run - `run_agent` parameters - Examples - [Agent frameworks > Mistral Agents](https://www.union.ai/docs/v2/flyte/integrations/agents/mistral.md) - Installation - Quick start - How it maps to Flyte - Driving a pre-created agent - Memory - Bounding a run - `run_agent` parameters - Examples - [Agent frameworks > LangChain](https://www.union.ai/docs/v2/flyte/integrations/agents/langchain.md) - Installation - Quick start - How it maps to Flyte - Pass a model instance, not a string - Bring your own agent - Memory - `run_agent` parameters - Examples - [Agent frameworks > LangGraph](https://www.union.ai/docs/v2/flyte/integrations/agents/langgraph.md) - Installation - Quick start - The node factories - How it maps to Flyte - Skipping the graph - Custom state - Memory - `run_agent` parameters - Examples - [Agent frameworks > Deep Agents](https://www.union.ai/docs/v2/flyte/integrations/agents/deepagents.md) - Installation - Quick start - How it maps to Flyte - Bring your own agent - Memory - Composing with Flyte - `run_agent` parameters - Examples - [Agent frameworks > CrewAI](https://www.union.ai/docs/v2/flyte/integrations/agents/crewai.md) - Installation - Quick start - How it maps to Flyte - Bring your own agent - Instructions and the built agent - Memory - `run_agent` parameters - Examples - [Agent frameworks > Pydantic AI](https://www.union.ai/docs/v2/flyte/integrations/agents/pydantic-ai.md) - Installation - Quick start - How it maps to Flyte - Bring your own agent - Memory - `run_agent` parameters - Examples - [Agent frameworks > Hermes](https://www.union.ai/docs/v2/flyte/integrations/agents/hermes.md) - Installation - Quick start - Credentials - How it maps to Flyte - Bring your own agent - Memory - `run_agent` parameters - Examples - [Software development tools](https://www.union.ai/docs/v2/flyte/integrations/software-development-tools.md) - Supported products - Quick start - The event model - Launch one run per event - Restrict which events launch runs - Test without an account - Deploy - Calling the product's API - Related features - Next steps - [Software development tools > GitHub](https://www.union.ai/docs/v2/flyte/integrations/software-development-tools/github.md) - Installation - The receiver - Set up the webhook in GitHub - Events - Scope and deduplication - The tasks it launches - Human review gates - GitHub App tokens - Test without a GitHub account - Examples - See also - [Software development tools > Slack](https://www.union.ai/docs/v2/flyte/integrations/software-development-tools/slack.md) - Installation - Credentials - The receiver - Delivery types - Set up the app in Slack - Events - Scope and deduplication - Send messages from tasks - Approvals - Test without a Slack workspace - Examples - See also - [Software development tools > Linear](https://www.union.ai/docs/v2/flyte/integrations/software-development-tools/linear.md) - Installation - The receiver - Set up the webhook in Linear - Events - Scope and deduplication - The task it launches - Test without a Linear workspace - Examples - See also - [Software development tools > ClickUp](https://www.union.ai/docs/v2/flyte/integrations/software-development-tools/clickup.md) - Installation - The receiver - Set up the webhook in ClickUp - Events - Scope and deduplication - The task it launches - Test without a ClickUp workspace - Examples - See also - [Software development tools > Jira](https://www.union.ai/docs/v2/flyte/integrations/software-development-tools/jira.md) - Installation - Authentication - The receiver - Set up the webhook in Jira - Events - Scope and deduplication - The task it launches - Test without a Jira site - Examples - See also - [Software development tools > GitLab](https://www.union.ai/docs/v2/flyte/integrations/software-development-tools/gitlab.md) - Installation - Authentication - The receiver - Set up the webhook in GitLab - Events - Scope and deduplication - Test without a GitLab account - See also - [BigQuery](https://www.union.ai/docs/v2/flyte/integrations/bigquery.md) - Installation - Quick start - Configuration - `BigQueryConfig` parameters - `BigQueryTask` parameters - Authentication - Query templating - Supported input types - Parameterized query example - Retrieving query results - API reference - [Code generation](https://www.union.ai/docs/v2/flyte/integrations/codegen.md) - Installation - Quick start - Two execution backends - LiteLLM (default) - Agent (Claude) - Providing data - Sample data - Schema and constraints - Inputs and outputs - Running generated code - One-shot execution with `result.run()` - Reusable task with `result.as_task()` - Error diagnosis - Durable execution - Replay logs - Caching - Non-determinism in agent mode - Observability - LiteLLM backend - Agent backend - Examples - Processing CSVs with different schemas - DataFrame analysis with constraints - Agent mode - Configuration - LiteLLM parameters - Image configuration - Skipping tests - Base packages - Best practices - API reference - `AutoCoderAgent` constructor - `generate()` parameters - `CodeGenEvalResult` fields - `CodeGenEvalResult` methods - [Dask](https://www.union.ai/docs/v2/flyte/integrations/dask.md) - When to use this plugin - Installation - Configuration - `Dask` parameters - `Scheduler` parameters - `WorkerGroup` parameters - Accessing the Dask client - Example - API reference - [Databricks](https://www.union.ai/docs/v2/flyte/integrations/databricks.md) - Installation - Quick start - Configuration - Spark fields (inherited) - Databricks-specific fields - `databricks_conf` structure - Authentication - Accessing the Spark session - API reference - [DBT](https://www.union.ai/docs/v2/flyte/integrations/dbt.md) - Installation - Quick start - Commands and arguments - Chaining dbt tasks - Outputs - Reports and failures - Custom dbt callbacks - Reference - `DbtTask` - Invocation inputs - [Grafana Agent Observability](https://www.union.ai/docs/v2/flyte/integrations/grafana-agent-observability.md) - Installation - What you get beyond agento11y on its own - Generations nest inside the task span - Flyte identity is bound onto agento11y's context - Durability is preserved end to end - Framework coverage - Configuration - `init()` parameters - Linking back from Grafana - Recording generations by hand - Content capture - Instrumenting a different backend - Limitations - Related - [Hugging Face](https://www.union.ai/docs/v2/flyte/integrations/huggingface.md) - When to use this plugin - Installation - What the plugin registers - Referencing a dataset on the Hub - It is a value, not just a default - Configs and splits - Config resolution - Omitting `split` concatenates everything - Reusing downloads across runs - Two caches, keyed differently - Reading only the columns you need - `Dataset` or `IterableDataset` - Passing datasets between tasks - Private and gated datasets - End-to-end: fine-tuning on IMDB - Running the examples - Common use cases - Hugging Face buckets as raw data storage - [Hydra](https://www.union.ai/docs/v2/flyte/integrations/hydra.md) - Installation - Requirements on tasks - A walkthrough config - Execution mode - Hydra launcher (`@hydra.main` scripts) - Python SDK - Single run - Grid sweep - Custom sweepers - Forwarding `flyte.with_runcontext` options - Flyte CLI (`flyte hydra run`) - Single run - Grid sweep - App-level vs Hydra-namespace overrides - `--follow` and `--no-wait` - Shell completion - Override grammar - Sweeps - Grid sweeps (BasicSweeper) - Bayesian / TPE sweeps (Optuna) - Sweep output directories - Task environment overrides - Prebuilt images - Applying overrides to child tasks - Renaming the task-env key - What `task_env` should not model - Structured configs (without YAML) - [JSONL](https://www.union.ai/docs/v2/flyte/integrations/jsonl.md) - When to use this plugin - Installation - Working with `JsonlFile` - Bulk and pre-serialized writes - Working with `JsonlDir` - Configuration and options - Compression - Write buffering - Prefetch - Error handling on read - Arrow batches - Common use cases - API reference - [Lance](https://www.union.ai/docs/v2/flyte/integrations/lance.md) - Lance vs. Parquet - Which one should you use? - How much does it matter? - Installation - What the plugin registers - Other dataframe types - Passing a dataset between tasks - Return a live handle - Hand off by reference - Choosing between them - Arrow tables and the explicit `lance` format - Reading only what you need - Multimodal data - Streaming a shuffled epoch - Reading the labels without the images - Parallel reads across fragments - Indices and schema evolution - Object storage and credentials - Measuring the difference - Running the examples - Common use cases - [MLflow](https://www.union.ai/docs/v2/flyte/integrations/mlflow.md) - Installation - Quick start - Autologging - Generic autologging - Framework-specific autologging - Run modes - Sharing a run across tasks - Creating independent runs - Nested runs - Workflow-level configuration - Per-task overrides - Configuration priority - Distributed training - MLflow UI links - Setup - Custom URL templates - Explicit links - Link behavior by run mode - Automatic Flyte tags - API reference - `mlflow_run` and `mlflow_config` - `get_mlflow_run` - `get_mlflow_context` - `Mlflow` - [NVIDIA Nsight Systems](https://www.union.ai/docs/v2/flyte/integrations/nsight.md) - Installation - Quick start - Reading the profile - Profiling part of a task - Labeling the timeline with NVTX - Choosing what to trace - Profiling distributed training - Turning profiling off - Permissions - Troubleshooting - How it works - Reference - [OmegaConf](https://www.union.ai/docs/v2/flyte/integrations/omegaconf.md) - Installation - Quick start - When to use this plugin - Building a DictConfig - From a plain dict - From a YAML file - From a dataclass (structured config) - From a base config plus overrides - Variable interpolation - Nested and deeply structured configs - DictConfigs that contain lists - ListConfig as input and output - Lists of primitives - Building a schedule from another task - Nested lists (list of lists) - Lists of DictConfigs - Lists of dataclass instances - Structured configs - Basic structured config - Schema reconstruction in the receiving task - Required (`MISSING`) fields - Advanced field types - Merging overrides on top of a structured base - Embedding rich Python values inside a plain DictConfig - Reserved-looking keys - YAML reports - Wire format - End-to-end example - [OpenTelemetry](https://www.union.ai/docs/v2/flyte/integrations/opentelemetry.md) - Installation - Quick start - What becomes a span - Span attributes - Your own spans - What's next - [OpenTelemetry > Exporters and configuration](https://www.union.ai/docs/v2/flyte/integrations/opentelemetry/configuration.md) - Environment variables - Configuring in code - `init()` parameters - Choosing an exporter - gRPC - Batching - Adopting a tracer provider you already have - Grafana Cloud - Linking back from Grafana - Shutting down - When nothing is configured - [OpenTelemetry > Traces across crashes and resumes](https://www.union.ai/docs/v2/flyte/integrations/opentelemetry/durable-traces.md) - Trace context, in and out - Inbound: joining a trace that started outside Flyte - Outbound: nested tasks, in other pods - One run, one trace - Replayed steps - A worked crash and resume - Flyte's own control-plane spans - Limitations - [Pandera](https://www.union.ai/docs/v2/flyte/integrations/pandera.md) - When to use this plugin - Installation - pandas - Polars - PySpark SQL - Defining schemas - Using schemas in tasks - Error handling with `ValidationConfig` - Image configuration - Pandas - Polars - PySpark SQL - Polars lazy frames - Examples - pandas - Polars - PySpark SQL - [Papermill](https://www.union.ai/docs/v2/flyte/integrations/papermill.md) - When to use this plugin - Installation - Quick start - Notebook setup - `parameters` cell - `outputs` cell - Inputs and outputs - Supported input types - Complex types: File, Dir, DataFrame - Outputs: single, multiple, none - Calling Flyte tasks from notebooks - Workflow patterns - Chaining notebooks - Mixing notebooks with regular tasks - Inline definition - Calling from sync vs. async tasks - Running a NotebookTask directly as the entrypoint - Reports and notebook artifacts - HTML report (default) - Notebook artifacts - Clean reports - Failure reports - Spark notebooks - Local testing - Execution options - `NotebookTask` reference - Helper functions - [Polars](https://www.union.ai/docs/v2/flyte/integrations/polars.md) - When to use this plugin - Installation - Using Polars DataFrames - Using LazyFrames - Interoperating with `flyte.io.DataFrame` - Common use cases - API reference - [PyTorch](https://www.union.ai/docs/v2/flyte/integrations/pytorch.md) - When to use this plugin - Installation - Configuration - `Elastic` parameters - `RunPolicy` parameters - NCCL tuning parameters - Writing a distributed training task - Example - API reference - [Ray](https://www.union.ai/docs/v2/flyte/integrations/ray.md) - When to use this plugin - Installation - Configuration - `RayJobConfig` parameters - `WorkerNodeConfig` parameters - `HeadNodeConfig` parameters - `AutoscalerOptionsConfig` parameters - Connecting to an existing cluster - Examples - API reference - [Slurm](https://www.union.ai/docs/v2/flyte/integrations/slurm.md) - Installation - Task types - Quick start - Running an existing sbatch script - Outputs from a script task - Configuration - Scheduling - Container and execution - Connection - Container runtimes - Container images - Object storage from inside the job - Outputs and caching - Output types - Caching - Job state mapping - Inspecting a job - Deployment - Connector image - Verifying the connector - Credentials - Network - Known gaps - Execution - Data and I/O - Operations - API reference - [Snowflake](https://www.union.ai/docs/v2/flyte/integrations/snowflake.md) - Installation - Quick start - Configuration - Required fields - Additional connection parameters - Authentication - Key-pair authentication - Password authentication - OAuth authentication - Query templating - Supported input types - Batched `INSERT` with list inputs - Parameterized `SELECT` - Multiple inputs - Retrieving query results - End-to-end example - [Spark](https://www.union.ai/docs/v2/flyte/integrations/spark.md) - When to use this plugin - Installation - Configuration - `Spark` parameters - Pod templates - Accessing the Spark session - Overriding configuration at runtime - Example - API reference - [TypeSafe AI](https://www.union.ai/docs/v2/flyte/integrations/typesafe-ai.md) - When to use this plugin - The three answer types - Setup - Declaring questions - Three ways to ask - Branching on the answers - Answers as task inputs and outputs - Two kinds of parallelism - A note on `IntEnum` - Next steps - [Weights & Biases](https://www.union.ai/docs/v2/flyte/integrations/wandb.md) - Installation - Quick start - What's next - [Weights & Biases > Experiments](https://www.union.ai/docs/v2/flyte/integrations/wandb/experiments.md) - Basic usage - Accessing the run object - Parent-child task relationships - Run modes - Using `run_mode="new"` for independent runs - Using `run_mode="shared"` for explicit sharing - Configuration with `wandb_config` - Workflow-level configuration - Overriding configuration for child tasks - Using traces with W&B runs - [Weights & Biases > Distributed training](https://www.union.ai/docs/v2/flyte/integrations/wandb/distributed_training.md) - Quick start - Run modes in distributed training - Single-node behavior - Multi-node behavior - Choosing run mode and rank scope - Single-node multi-GPU - Basic example with `auto` mode - Using `shared` mode for per-rank metrics - Using `new` mode for per-rank runs - Multi-node training with `Elastic` - Global scope (default): Single run across all nodes - Worker scope: One run per node - Shared mode: All ranks log to the same run - New mode: Separate run per rank - How it works - Run ID patterns - [Weights & Biases > Sweeps](https://www.union.ai/docs/v2/flyte/integrations/wandb/sweeps.md) - Creating a sweep - Running parallel agents - Writing objective functions - [Weights & Biases > Downloading logs](https://www.union.ai/docs/v2/flyte/integrations/wandb/downloading_logs.md) - Automatic download - Accessing run directories during execution - [Weights & Biases > Constraints and best practices](https://www.union.ai/docs/v2/flyte/integrations/wandb/constraints_and_best_practices.md) - Decorator ordering - Traces cannot use decorators - Maximum sweep agents - Configuration priority - Run ID generation - Sync delay for local files - Shared run mode requirements - Objective functions for sweeps - Error handling - [Weights & Biases > Manual integration](https://www.union.ai/docs/v2/flyte/integrations/wandb/manual.md) - Using the Wandb link class - With a custom run ID - Adding links at runtime with override - Using the `WandbSweep` link class --- ## Reference - [Flyte CLI](https://www.union.ai/docs/v2/flyte/api-reference/flyte-cli.md) - flyte - flyte abort - flyte build - flyte create - flyte delete - flyte deploy - flyte edit - flyte gen - flyte get - flyte prefetch - flyte proxy - flyte rerun - flyte run - flyte serve - flyte signal - flyte start - flyte stop - flyte update - flyte whoami - [Flyte SDK](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk.md) - Directory - Classes - Protocols - Functions - Packages - [Flyte SDK > flyte](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.md) - Directory - Classes - Protocols - Methods - Variables - Methods - [Flyte SDK > flyte > AppHandle](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/apphandle.md) - Properties - Methods - activate() - deactivate() - ephemeral_ctx() - ephemeral_ctx_sync() - is_active() - is_deactivated() - [Flyte SDK > flyte > AsyncFunctionTaskTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/asyncfunctiontasktemplate.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte > Backoff](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/backoff.md) - Parameters - Methods - compute_delay() - [Flyte SDK > flyte > BaseCheckpoint](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/basecheckpoint.md) - Properties - Methods - load() - load_sync() - prev_exists() - save() - save_sync() - [Flyte SDK > flyte > Cache](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/cache.md) - Parameters - Methods - get_ignored_inputs() - get_version() - is_enabled() - [Flyte SDK > flyte > CachePolicy](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/cachepolicy.md) - Methods - get_version() - [Flyte SDK > flyte > Checkpoint](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/checkpoint.md) - Parameters - Properties - Methods - load() - load_sync() - prev_exists() - save() - save_sync() - [Flyte SDK > flyte > ConditionWebhook](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/conditionwebhook.md) - Parameters - [Flyte SDK > flyte > Cron](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/cron.md) - Parameters - Properties - [Flyte SDK > flyte > Device](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/device.md) - Parameters - [Flyte SDK > flyte > Documentation](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/documentation.md) - Parameters - [Flyte SDK > flyte > Environment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/environment.md) - Parameters - Methods - add_dependency() - clone_with() - [Flyte SDK > flyte > FixedRate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/fixedrate.md) - Parameters - [Flyte SDK > flyte > Image](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/image.md) - Parameters - Properties - Methods - clone() - from_base() - from_debian_base() - from_dockerfile() - from_pixi_script() - from_ref_name() - from_uv_script() - validate() - with_apt_packages() - with_code_bundle() - with_commands() - with_dockerignore() - with_env_vars() - with_local_rs_controller() - with_local_v2() - with_local_v2_plugins() - with_pip_packages() - with_pixi_project() - with_poetry_project() - with_requirements() - with_source_file() - with_source_folder() - with_uv_project() - with_workdir() - [Flyte SDK > flyte > ImageBuild](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/imagebuild.md) - Parameters - [Flyte SDK > flyte > Link](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/link.md) - Methods - get_link() - [Flyte SDK > flyte > OnArtifact](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/onartifact.md) - Parameters - [Flyte SDK > flyte > PodTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/podtemplate.md) - Parameters - Methods - allow_fuse() - allow_nested_sandboxing() - from_spec() - to_k8s_pod() - with_termination_grace_period() - [Flyte SDK > flyte > Resources](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/resources.md) - Parameters - Methods - get_device() - get_shared_memory() - [Flyte SDK > flyte > RetryStrategy](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/retrystrategy.md) - Parameters - [Flyte SDK > flyte > ReusePolicy](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/reusepolicy.md) - Parameters - Properties - Methods - get_scaledown_ttl() - [Flyte SDK > flyte > Secret](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/secret.md) - Parameters - Methods - stable_hash() - [Flyte SDK > flyte > TaskEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/taskenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - from_task() - task() - [Flyte SDK > flyte > TaskTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/tasktemplate.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte > Timeout](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/timeout.md) - Parameters - [Flyte SDK > flyte > Trigger](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/trigger.md) - Parameters - Methods - daily() - hourly() - minutely() - monthly() - weekly() - [Flyte SDK > flyte > TriggeredPartition](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte/triggeredpartition.md) - Parameters - [Flyte SDK > flyte.ai.agents](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.md) - Directory - Classes - Protocols - Errors - Methods - Variables - Methods - [Flyte SDK > flyte.ai.agents > AccessDenied](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/accessdenied.md) - [Flyte SDK > flyte.ai.agents > Agent](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/agent.md) - Parameters - Properties - Methods - add_tool() - approval_callback() - call_llm() - run() - tool_descriptions() - [Flyte SDK > flyte.ai.agents > AgentEvent](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/agentevent.md) - Parameters - [Flyte SDK > flyte.ai.agents > AgentProtocol](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/agentprotocol.md) - Methods - run() - tool_descriptions() - [Flyte SDK > flyte.ai.agents > AgentResult](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/agentresult.md) - Parameters - [Flyte SDK > flyte.ai.agents > AgentTool](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/agenttool.md) - Parameters - Methods - aio() - to_openai_format() - [Flyte SDK > flyte.ai.agents > ConcurrencyError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/concurrencyerror.md) - Parameters - [Flyte SDK > flyte.ai.agents > LLMMessage](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/llmmessage.md) - Parameters - [Flyte SDK > flyte.ai.agents > MCPServerSpec](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/mcpserverspec.md) - Parameters - [Flyte SDK > flyte.ai.agents > MemoryMeta](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/memorymeta.md) - Parameters - [Flyte SDK > flyte.ai.agents > MemoryStore](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/memorystore.md) - Parameters - Methods - append() - audit_tail() - audit_tail_sync() - create() - current_sha() - exists() - extend() - flush_messages() - flush_messages_sync() - get_meta() - get_or_create() - list_paths() - read_json() - read_text() - remote_path_for_key() - save() - write_json() - write_text() - [Flyte SDK > flyte.ai.agents > MemoryStoreError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/memorystoreerror.md) - [Flyte SDK > flyte.ai.agents > ToolFn](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents/toolfn.md) - Parameters - [Flyte SDK > flyte.ai.agents.agent](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.agent.md) - Directory - Classes - Variables - [Flyte SDK > flyte.ai.agents.agent > Agent](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.agent/agent.md) - Parameters - Properties - Methods - add_tool() - approval_callback() - call_llm() - run() - tool_descriptions() - [Flyte SDK > flyte.ai.agents.agent > AgentEvent](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.agent/agentevent.md) - Parameters - [Flyte SDK > flyte.ai.agents.memory](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.memory.md) - Directory - Classes - Errors - Variables - [Flyte SDK > flyte.ai.agents.memory > AccessDenied](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.memory/accessdenied.md) - [Flyte SDK > flyte.ai.agents.memory > ConcurrencyError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.memory/concurrencyerror.md) - Parameters - [Flyte SDK > flyte.ai.agents.memory > MemoryMeta](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.memory/memorymeta.md) - Parameters - [Flyte SDK > flyte.ai.agents.memory > MemoryStore](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.memory/memorystore.md) - Parameters - Methods - append() - audit_tail() - audit_tail_sync() - create() - current_sha() - exists() - extend() - flush_messages() - flush_messages_sync() - get_meta() - get_or_create() - list_paths() - read_json() - read_text() - remote_path_for_key() - save() - write_json() - write_text() - [Flyte SDK > flyte.ai.agents.memory > MemoryStoreError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.memory/memorystoreerror.md) - [Flyte SDK > flyte.ai.agents.protocol](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.protocol.md) - Directory - Classes - Protocols - Variables - [Flyte SDK > flyte.ai.agents.protocol > AgentProtocol](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.protocol/agentprotocol.md) - Methods - run() - tool_descriptions() - [Flyte SDK > flyte.ai.agents.protocol > AgentResult](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.agents.protocol/agentresult.md) - Parameters - [Flyte SDK > flyte.ai.chat](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.chat.md) - Directory - Classes - Methods - Variables - Methods - [Flyte SDK > flyte.ai.chat > AgentChatAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.chat/agentchatappenvironment.md) - Parameters - Properties - Methods - add_dependency() - build_fastapi_app() - clone_with() - container_args() - container_cmd() - container_command() - get_port() - on_shutdown() - on_startup() - server() - [Flyte SDK > flyte.ai.chat > CustomTheme](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.chat/customtheme.md) - Parameters - Methods - to_css() - [Flyte SDK > flyte.ai.chat.app](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.chat.app.md) - Directory - Classes - Variables - [Flyte SDK > flyte.ai.chat.app > AgentChatAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.chat.app/agentchatappenvironment.md) - Parameters - Properties - Methods - add_dependency() - build_fastapi_app() - clone_with() - container_args() - container_cmd() - container_command() - get_port() - on_shutdown() - on_startup() - server() - [Flyte SDK > flyte.ai.chat.app > CustomTheme](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.chat.app/customtheme.md) - Parameters - Methods - to_css() - [Flyte SDK > flyte.ai.mcp](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.mcp.md) - Directory - Classes - Methods - Variables - Methods - [Flyte SDK > flyte.ai.mcp > FlyteMCPAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.mcp/flytemcpappenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - container_command() - get_port() - on_shutdown() - on_startup() - resolve_scope() - resolved_allowed_hosts() - resolved_allowed_origins() - run_stdio() - run_stdio_async() - server() - [Flyte SDK > flyte.ai.mcp > MCPAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.ai.mcp/mcpappenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - container_command() - get_port() - on_shutdown() - on_startup() - run_stdio() - run_stdio_async() - server() - [Flyte SDK > flyte.app](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.app > AppEndpoint](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/appendpoint.md) - Parameters - Methods - check_type() - get() - materialize() - [Flyte SDK > flyte.app > AppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/appenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - get_port() - on_shutdown() - on_startup() - server() - [Flyte SDK > flyte.app > ArtifactValue](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/artifactvalue.md) - Parameters - Properties - Methods - check_type() - get() - materialize() - model_post_init() - [Flyte SDK > flyte.app > ConnectorEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/connectorenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - get_port() - on_shutdown() - on_startup() - server() - [Flyte SDK > flyte.app > DeployedAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/deployedappenvironment.md) - Parameters - Methods - env_repr() - get_name() - summary_repr() - table_repr() - [Flyte SDK > flyte.app > Domain](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/domain.md) - Parameters - [Flyte SDK > flyte.app > Link](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/link.md) - Parameters - [Flyte SDK > flyte.app > Parameter](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/parameter.md) - Parameters - [Flyte SDK > flyte.app > Port](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/port.md) - Parameters - [Flyte SDK > flyte.app > RunOutput](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/runoutput.md) - Parameters - Methods - check_type() - get() - materialize() - [Flyte SDK > flyte.app > Scaling](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/scaling.md) - Parameters - Methods - get_replicas() - [Flyte SDK > flyte.app > Subdomain](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/subdomain.md) - Parameters - Methods - from_app_name() - from_function() - resolve() - [Flyte SDK > flyte.app > Timeouts](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app/timeouts.md) - Parameters - [Flyte SDK > flyte.app.extras](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app.extras.md) - Directory - Classes - [Flyte SDK > flyte.app.extras > FastAPIAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app.extras/fastapiappenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - container_command() - get_port() - on_shutdown() - on_startup() - server() - [Flyte SDK > flyte.app.extras > FastAPIPassthroughAuthMiddleware](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app.extras/fastapipassthroughauthmiddleware.md) - Parameters - Methods - dispatch() - extract_authorization_header() - extract_cookie_header() - extract_custom_header() - [Flyte SDK > flyte.app.extras > FlyteWebhookAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.app.extras/flytewebhookappenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - container_command() - get_port() - on_shutdown() - on_startup() - server() - [Flyte SDK > flyte.artifacts](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.artifacts.md) - Directory - Classes - Protocols - Methods - Variables - Methods - [Flyte SDK > flyte.artifacts > Artifact](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.artifacts/artifact.md) - Methods - get_artifact_metadata() - [Flyte SDK > flyte.artifacts > ArtifactKey](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.artifacts/artifactkey.md) - [Flyte SDK > flyte.artifacts > ArtifactVersionId](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.artifacts/artifactversionid.md) - [Flyte SDK > flyte.artifacts > Card](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.artifacts/card.md) - Parameters - Methods - create_from() - [Flyte SDK > flyte.artifacts > Metadata](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.artifacts/metadata.md) - Parameters - Methods - create_model_metadata() - [Flyte SDK > flyte.artifacts > TimePartition](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.artifacts/timepartition.md) - Parameters - [Flyte SDK > flyte.clustered](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.clustered.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.clustered > ClusteredTaskEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.clustered/clusteredtaskenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - from_task() - task() - to_custom_dict() - [Flyte SDK > flyte.clustered > ClusteredTaskTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.clustered/clusteredtasktemplate.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte.clustered > ClusterFailurePolicy](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.clustered/clusterfailurepolicy.md) - Parameters - [Flyte SDK > flyte.clustered > JaxRun](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.clustered/jaxrun.md) - Parameters - [Flyte SDK > flyte.clustered > TorchRun](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.clustered/torchrun.md) - Parameters - [Flyte SDK > flyte.config](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.config.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.config > Config](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.config/config.md) - Parameters - Methods - auto() - with_params() - [Flyte SDK > flyte.connectors](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.connectors.md) - Directory - Classes - [Flyte SDK > flyte.connectors > AsyncConnector](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.connectors/asyncconnector.md) - Methods - create() - delete() - get() - get_logs() - get_metrics() - [Flyte SDK > flyte.connectors > AsyncConnectorExecutorMixin](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.connectors/asyncconnectorexecutormixin.md) - Methods - execute() - [Flyte SDK > flyte.connectors > ConnectorRegistry](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.connectors/connectorregistry.md) - Methods - get_connector() - register() - [Flyte SDK > flyte.connectors > ConnectorService](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.connectors/connectorservice.md) - Methods - run() - [Flyte SDK > flyte.connectors > Resource](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.connectors/resource.md) - Parameters - [Flyte SDK > flyte.connectors > ResourceMeta](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.connectors/resourcemeta.md) - Parameters - Methods - decode() - encode() - [Flyte SDK > flyte.connectors.utils](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.connectors.utils.md) - Directory - Methods - Methods - [Flyte SDK > flyte.durable](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.durable.md) - Directory - Methods - Methods - [Flyte SDK > flyte.errors](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors.md) - Directory - Errors - Methods - Variables - Methods - [Flyte SDK > flyte.errors > ActionAbortedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/actionabortederror.md) - Parameters - [Flyte SDK > flyte.errors > ActionNotFoundError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/actionnotfounderror.md) - [Flyte SDK > flyte.errors > AppleSiliconBridgeUnreachableError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/applesiliconbridgeunreachableerror.md) - Parameters - [Flyte SDK > flyte.errors > AppleSiliconDaemonOOMError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/applesilicondaemonoomerror.md) - Parameters - [Flyte SDK > flyte.errors > AppleSiliconDaemonStartFailedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/applesilicondaemonstartfailederror.md) - Parameters - [Flyte SDK > flyte.errors > AppleSiliconError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/applesiliconerror.md) - Parameters - [Flyte SDK > flyte.errors > AppleSiliconHostExecFailedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/applesiliconhostexecfailederror.md) - Parameters - [Flyte SDK > flyte.errors > AppleSiliconModelNotFoundError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/applesiliconmodelnotfounderror.md) - Parameters - [Flyte SDK > flyte.errors > AppleSiliconNotAvailableError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/applesiliconnotavailableerror.md) - Parameters - [Flyte SDK > flyte.errors > BaseRuntimeError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/baseruntimeerror.md) - Parameters - [Flyte SDK > flyte.errors > CodeBundleError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/codebundleerror.md) - Parameters - [Flyte SDK > flyte.errors > ConditionAlreadyExistsError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/conditionalreadyexistserror.md) - Parameters - [Flyte SDK > flyte.errors > ConditionFailedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/conditionfailederror.md) - Parameters - [Flyte SDK > flyte.errors > ConditionNotFoundError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/conditionnotfounderror.md) - Parameters - [Flyte SDK > flyte.errors > ConditionTimedoutError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/conditiontimedouterror.md) - Parameters - [Flyte SDK > flyte.errors > CustomError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/customerror.md) - Parameters - Methods - from_exception() - [Flyte SDK > flyte.errors > DeadlineExceededError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/deadlineexceedederror.md) - Parameters - [Flyte SDK > flyte.errors > DeploymentError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/deploymenterror.md) - Parameters - [Flyte SDK > flyte.errors > GPUFaultError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/gpufaulterror.md) - Parameters - Properties - [Flyte SDK > flyte.errors > GPUFaultSystemError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/gpufaultsystemerror.md) - Parameters - Properties - [Flyte SDK > flyte.errors > GPUFaultUserError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/gpufaultusererror.md) - Parameters - Properties - [Flyte SDK > flyte.errors > ImageBuildError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/imagebuilderror.md) - Parameters - [Flyte SDK > flyte.errors > ImagePullBackOffError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/imagepullbackofferror.md) - Parameters - [Flyte SDK > flyte.errors > InitializationError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/initializationerror.md) - Parameters - [Flyte SDK > flyte.errors > InlineIOMaxBytesBreached](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/inlineiomaxbytesbreached.md) - Parameters - [Flyte SDK > flyte.errors > InvalidImageNameError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/invalidimagenameerror.md) - Parameters - [Flyte SDK > flyte.errors > InvalidPackageError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/invalidpackageerror.md) - Parameters - [Flyte SDK > flyte.errors > LogsNotYetAvailableError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/logsnotyetavailableerror.md) - Parameters - [Flyte SDK > flyte.errors > MaxQueuedTimeExceededError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/maxqueuedtimeexceedederror.md) - Parameters - [Flyte SDK > flyte.errors > MaxRuntimeExceededError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/maxruntimeexceedederror.md) - Parameters - [Flyte SDK > flyte.errors > ModuleLoadError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/moduleloaderror.md) - Parameters - [Flyte SDK > flyte.errors > NonRecoverableError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/nonrecoverableerror.md) - Parameters - [Flyte SDK > flyte.errors > NotInTaskContextError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/notintaskcontexterror.md) - Parameters - [Flyte SDK > flyte.errors > OnlyAsyncIOSupportedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/onlyasynciosupportederror.md) - Parameters - [Flyte SDK > flyte.errors > OOMError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/oomerror.md) - Parameters - [Flyte SDK > flyte.errors > ParameterMaterializationError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/parametermaterializationerror.md) - Parameters - [Flyte SDK > flyte.errors > PrimaryContainerNotFoundError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/primarycontainernotfounderror.md) - Parameters - [Flyte SDK > flyte.errors > RemoteTaskNotFoundError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/remotetasknotfounderror.md) - Parameters - [Flyte SDK > flyte.errors > RemoteTaskUsageError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/remotetaskusageerror.md) - Parameters - [Flyte SDK > flyte.errors > ResourceExhaustedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/resourceexhaustederror.md) - Parameters - [Flyte SDK > flyte.errors > RestrictedTypeError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/restrictedtypeerror.md) - Parameters - [Flyte SDK > flyte.errors > RetriesExhaustedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/retriesexhaustederror.md) - Parameters - [Flyte SDK > flyte.errors > RuntimeDataValidationError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/runtimedatavalidationerror.md) - Parameters - [Flyte SDK > flyte.errors > RuntimeSystemError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/runtimesystemerror.md) - Parameters - [Flyte SDK > flyte.errors > RuntimeUnknownError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/runtimeunknownerror.md) - Parameters - [Flyte SDK > flyte.errors > RuntimeUserError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/runtimeusererror.md) - Parameters - [Flyte SDK > flyte.errors > SlowDownError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/slowdownerror.md) - Parameters - [Flyte SDK > flyte.errors > SyncTaskCallInAsyncContextError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/synctaskcallinasynccontexterror.md) - Parameters - [Flyte SDK > flyte.errors > TaskInterruptedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/taskinterruptederror.md) - Parameters - [Flyte SDK > flyte.errors > TaskTimeoutError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/tasktimeouterror.md) - Parameters - [Flyte SDK > flyte.errors > TraceDoesNotAllowNestedTasksError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/tracedoesnotallownestedtaskserror.md) - Parameters - [Flyte SDK > flyte.errors > UnionRpcError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.errors/unionrpcerror.md) - Parameters - [Flyte SDK > flyte.extend](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extend.md) - Directory - Classes - Protocols - Methods - Variables - Methods - [Flyte SDK > flyte.extend > AsyncFunctionTaskTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extend/asyncfunctiontasktemplate.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte.extend > ImageBuildEngine](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extend/imagebuildengine.md) - [Flyte SDK > flyte.extend > ImageBuilder](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extend/imagebuilder.md) - Methods - build_image() - get_checkers() - [Flyte SDK > flyte.extend > ImageChecker](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extend/imagechecker.md) - Methods - image_exists() - [Flyte SDK > flyte.extend > RuntimeTaskTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extend/runtimetasktemplate.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte.extend > TaskTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extend/tasktemplate.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte.extras](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.md) - Directory - Classes - Protocols - Methods - Methods - [Flyte SDK > flyte.extras > BatchStats](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/batchstats.md) - Parameters - Properties - [Flyte SDK > flyte.extras > ContainerTask](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/containertask.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte.extras > CostEstimator](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/costestimator.md) - Methods - estimate_cost() - [Flyte SDK > flyte.extras > DynamicBatcher](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/dynamicbatcher.md) - Parameters - Properties - Methods - start() - stop() - submit() - submit_batch() - [Flyte SDK > flyte.extras > Prompt](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/prompt.md) - Parameters - Methods - estimate_tokens() - [Flyte SDK > flyte.extras > Sleep](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/sleep.md) - Parameters - [Flyte SDK > flyte.extras > SleepTask](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/sleeptask.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte.extras > TokenBatcher](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/tokenbatcher.md) - Parameters - Properties - Methods - start() - stop() - submit() - submit_batch() - [Flyte SDK > flyte.extras > TokenEstimator](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras/tokenestimator.md) - Methods - estimate_tokens() - [Flyte SDK > flyte.extras.model_streamer](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.model_streamer.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.extras.model_streamer > LoadResult](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.model_streamer/loadresult.md) - Parameters - [Flyte SDK > flyte.extras.model_streamer > ModelStreamer](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.model_streamer/modelstreamer.md) - Parameters - Methods - download_metadata() - load_into() - state_dict() - stream() - stream_sync() - [Flyte SDK > flyte.extras.shell](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.shell.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.extras.shell > FlagSpec](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.shell/flagspec.md) - Parameters - Methods - coerce() - [Flyte SDK > flyte.extras.shell > Glob](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.shell/glob.md) - Parameters - [Flyte SDK > flyte.extras.shell > Stderr](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.shell/stderr.md) - Parameters - [Flyte SDK > flyte.extras.shell > Stdout](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.shell/stdout.md) - Parameters - [Flyte SDK > flyte.extras.webhooks](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks.md) - Directory - Classes - Errors - Methods - Variables - Methods - [Flyte SDK > flyte.extras.webhooks > EventType](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks/eventtype.md) - Parameters - [Flyte SDK > flyte.extras.webhooks > Provider](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks/provider.md) - Parameters - [Flyte SDK > flyte.extras.webhooks > RunOnceResult](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks/runonceresult.md) - [Flyte SDK > flyte.extras.webhooks > SignatureError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks/signatureerror.md) - [Flyte SDK > flyte.extras.webhooks > WebhookAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks/webhookappenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - container_command() - get_port() - on_event() - on_shutdown() - on_startup() - server() - [Flyte SDK > flyte.extras.webhooks > WebhookEvent](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks/webhookevent.md) - Parameters - Properties - Methods - dedupe_key() - [Flyte SDK > flyte.extras.webhooks > WebhookPluginError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks/webhookpluginerror.md) - [Flyte SDK > flyte.extras.webhooks.testing](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks.testing.md) - Directory - Protocols - Methods - Methods - [Flyte SDK > flyte.extras.webhooks.testing > ProviderFactory](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.extras.webhooks.testing/providerfactory.md) - [Flyte SDK > flyte.git](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.git.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.git > GitStatus](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.git/gitstatus.md) - Parameters - Methods - build_url() - from_current_repo() - [Flyte SDK > flyte.io](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io.md) - IO data types - Directory - Classes - Variables - [Flyte SDK > flyte.io > DataFrame](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io/dataframe.md) - Parameters - Properties - Methods - all() - all_sync() - column_names() - columns() - deserialize_dataframe() - from_df() - from_existing_remote() - from_local() - from_local_sync() - iter() - model_post_init() - open() - schema_match() - serialize_dataframe() - set_literal() - wrap_df() - [Flyte SDK > flyte.io > Dir](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io/dir.md) - Parameters - Properties - Methods - download() - download_sync() - empty() - exists() - exists_sync() - from_existing_remote() - from_local() - from_local_sync() - get_file() - get_file_sync() - list_files() - list_files_sync() - model_post_init() - new_remote() - walk() - walk_sync() - [Flyte SDK > flyte.io > EmptyDir](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io/emptydir.md) - Parameters - Properties - Methods - download() - download_sync() - empty() - exists() - exists_sync() - from_existing_remote() - from_local() - from_local_sync() - get_file() - get_file_sync() - list_files() - list_files_sync() - model_post_init() - new_remote() - walk() - walk_sync() - [Flyte SDK > flyte.io > File](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io/file.md) - Parameters - Properties - Methods - download() - download_sync() - exists() - exists_sync() - from_existing_remote() - from_local() - from_local_sync() - model_post_init() - named_remote() - new_remote() - open() - open_sync() - [Flyte SDK > flyte.io > HashFunction](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io/hashfunction.md) - Parameters - Methods - from_fn() - reset() - result() - update() - [Flyte SDK > flyte.io.extend](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io.extend.md) - Directory - Classes - [Flyte SDK > flyte.io.extend > DataFrameDecoder](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io.extend/dataframedecoder.md) - Parameters - Properties - Methods - decode() - [Flyte SDK > flyte.io.extend > DataFrameEncoder](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io.extend/dataframeencoder.md) - Parameters - Properties - Methods - encode() - [Flyte SDK > flyte.io.extend > DataFrameTransformerEngine](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.io.extend/dataframetransformerengine.md) - Parameters - Properties - Methods - assert_type() - encode() - from_binary_idl() - get_decoder() - get_encoder() - get_literal_type() - get_structured_dataset_type() - guess_python_type() - isinstance_generic() - iter_as() - open_as() - register() - register_for_protocol() - register_renderer() - schema_match() - to_html() - to_literal() - to_python_value() - [Flyte SDK > flyte.models](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models.md) - Directory - Classes - Methods - Variables - Methods - [Flyte SDK > flyte.models > ActionID](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/actionid.md) - Parameters - Methods - create_random() - new_sub_action() - new_sub_action_from() - unique_id_str() - [Flyte SDK > flyte.models > ActionPhase](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/actionphase.md) - Parameters - [Flyte SDK > flyte.models > CheckpointPaths](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/checkpointpaths.md) - Parameters - [Flyte SDK > flyte.models > CodeBundle](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/codebundle.md) - Parameters - Methods - with_downloaded_path() - [Flyte SDK > flyte.models > GroupData](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/groupdata.md) - Parameters - [Flyte SDK > flyte.models > NativeInterface](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/nativeinterface.md) - Parameters - Properties - Methods - convert_to_kwargs() - from_callable() - from_types() - get_input_types() - has_outputs() - num_required_inputs() - required_inputs() - [Flyte SDK > flyte.models > PathRewrite](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/pathrewrite.md) - Parameters - Methods - from_str() - [Flyte SDK > flyte.models > RawDataPath](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/rawdatapath.md) - Parameters - Methods - from_local_folder() - get_random_remote_path() - [Flyte SDK > flyte.models > SerializationContext](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/serializationcontext.md) - Parameters - Methods - get_entrypoint_path() - [Flyte SDK > flyte.models > TaskContext](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.models/taskcontext.md) - Parameters - Properties - Methods - is_in_cluster() - replace() - [Flyte SDK > flyte.notify](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.notify.md) - Directory - Classes - [Flyte SDK > flyte.notify > Email](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.notify/email.md) - Parameters - [Flyte SDK > flyte.notify > NamedDelivery](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.notify/nameddelivery.md) - Parameters - [Flyte SDK > flyte.notify > NamedRule](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.notify/namedrule.md) - Parameters - [Flyte SDK > flyte.notify > Notification](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.notify/notification.md) - Parameters - [Flyte SDK > flyte.notify > Slack](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.notify/slack.md) - Parameters - [Flyte SDK > flyte.notify > Teams](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.notify/teams.md) - Parameters - [Flyte SDK > flyte.notify > Webhook](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.notify/webhook.md) - Parameters - [Flyte SDK > flyte.prefetch](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.prefetch.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.prefetch > HuggingFaceModelInfo](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.prefetch/huggingfacemodelinfo.md) - Parameters - [Flyte SDK > flyte.prefetch > ShardConfig](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.prefetch/shardconfig.md) - Parameters - [Flyte SDK > flyte.prefetch > StoredModelInfo](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.prefetch/storedmodelinfo.md) - Parameters - [Flyte SDK > flyte.prefetch > VLLMShardArgs](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.prefetch/vllmshardargs.md) - Parameters - Methods - get_vllm_args() - [Flyte SDK > flyte.remote](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.remote > Action](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/action.md) - Parameters - Properties - Methods - abort() - details() - done() - download_code() - get() - get_logs() - get_report() - listall() - show_logs() - sync() - to_dict() - to_json() - wait() - watch() - [Flyte SDK > flyte.remote > ActionDetails](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/actiondetails.md) - Parameters - Properties - Methods - done() - get() - get_details() - get_phase_transitions() - input_literals() - inputs() - logs_available() - output_literals() - outputs() - to_dict() - to_json() - typed_inputs() - typed_outputs() - watch() - watch_updates() - [Flyte SDK > flyte.remote > ActionInputs](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/actioninputs.md) - Parameters - Methods - to_dict() - to_json() - [Flyte SDK > flyte.remote > ActionOutputs](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/actionoutputs.md) - Parameters - Properties - Methods - to_dict() - to_json() - [Flyte SDK > flyte.remote > App](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/app.md) - Parameters - Properties - Methods - activate() - create() - deactivate() - delete() - ephemeral_ctx() - ephemeral_ctx_sync() - get() - is_active() - is_deactivated() - listall() - replace() - show_logs() - to_dict() - to_json() - update() - watch() - [Flyte SDK > flyte.remote > Artifact](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/artifact.md) - Parameters - Properties - Methods - coerce_to_literal() - create() - declare() - delete() - get() - get_schema() - list_names() - listall() - partition_values() - to_dict() - to_json() - to_python() - [Flyte SDK > flyte.remote > Condition](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/condition.md) - Parameters - Properties - Methods - get() - listall() - signal() - to_dict() - to_json() - [Flyte SDK > flyte.remote > PartitionSchema](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/partitionschema.md) - Parameters - Properties - Methods - from_pb2() - to_pb2() - [Flyte SDK > flyte.remote > Project](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/project.md) - Parameters - Methods - archive() - create() - get() - listall() - to_dict() - to_json() - unarchive() - update() - [Flyte SDK > flyte.remote > Run](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/run.md) - Parameters - Properties - Methods - abort() - code_bundle() - details() - done() - download_code() - first_failure() - get() - get_debug_url() - get_logs() - get_report() - input_literals() - inputs() - listall() - output_literals() - outputs() - raise_for_status() - show_logs() - sync() - to_dict() - to_json() - typed_inputs() - typed_outputs() - wait() - watch() - [Flyte SDK > flyte.remote > RunDetails](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/rundetails.md) - Parameters - Properties - Methods - done() - get() - get_details() - input_literals() - inputs() - output_literals() - outputs() - to_dict() - to_json() - typed_inputs() - typed_outputs() - [Flyte SDK > flyte.remote > Secret](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/secret.md) - Parameters - Properties - Methods - create() - delete() - get() - listall() - to_dict() - to_json() - [Flyte SDK > flyte.remote > Settings](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/settings.md) - Parameters - Methods - available_keys() - effective_values() - get_settings_for_edit() - local_overrides() - parse_yaml() - scope_description() - to_dict() - to_json() - to_yaml() - to_yaml_sections() - update_settings() - [Flyte SDK > flyte.remote > Task](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/task.md) - Parameters - Properties - Methods - get() - listall() - to_dict() - to_json() - [Flyte SDK > flyte.remote > TaskDetails](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/taskdetails.md) - Parameters - Properties - Methods - fetch() - get() - override() - to_dict() - to_json() - [Flyte SDK > flyte.remote > TimeFilter](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/timefilter.md) - Parameters - [Flyte SDK > flyte.remote > Trigger](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/trigger.md) - Parameters - Properties - Methods - create() - delete() - get() - get_details() - listall() - to_dict() - to_json() - update() - [Flyte SDK > flyte.remote > TriggerDetails](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/triggerdetails.md) - Parameters - Properties - Methods - get() - to_dict() - to_json() - [Flyte SDK > flyte.remote > User](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.remote/user.md) - Parameters - Methods - get() - name() - subject() - to_dict() - to_json() - [Flyte SDK > flyte.report](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.report.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.report > Report](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.report/report.md) - Parameters - Methods - get_final_report() - get_tab() - has_content() - [Flyte SDK > flyte.report > Timeline](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.report/timeline.md) - Parameters - Methods - heading() - row() - [Flyte SDK > flyte.sandbox](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.sandbox.md) - Directory - Classes - Methods - Variables - Methods - [Flyte SDK > flyte.sandbox > CodeTaskTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.sandbox/codetasktemplate.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte.sandbox > ImageConfig](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.sandbox/imageconfig.md) - Parameters - [Flyte SDK > flyte.sandbox > SandboxedConfig](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.sandbox/sandboxedconfig.md) - Parameters - [Flyte SDK > flyte.sandbox > SandboxedTaskTemplate](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.sandbox/sandboxedtasktemplate.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Flyte SDK > flyte.storage](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.storage.md) - Directory - Classes - Methods - Methods - [Flyte SDK > flyte.storage > ABFS](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.storage/abfs.md) - Parameters - Methods - auto() - get_fsspec_kwargs() - [Flyte SDK > flyte.storage > GCS](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.storage/gcs.md) - Parameters - Methods - auto() - get_fsspec_kwargs() - [Flyte SDK > flyte.storage > S3](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.storage/s3.md) - Parameters - Methods - auto() - for_sandbox() - get_fsspec_kwargs() - [Flyte SDK > flyte.storage > Storage](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.storage/storage.md) - Parameters - Methods - auto() - get_fsspec_kwargs() - [Flyte SDK > flyte.syncify](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.syncify.md) - Creating a Syncify Instance - How does it work? - Directory - Classes - [Flyte SDK > flyte.syncify > Syncify](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.syncify/syncify.md) - Parameters - [Flyte SDK > flyte.types](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.types.md) - Directory - Classes - Protocols - Errors - Methods - Methods - [Flyte SDK > flyte.types > FlytePickle](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.types/flytepickle.md) - Methods - from_pickle() - python_type() - to_pickle() - [Flyte SDK > flyte.types > Renderable](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.types/renderable.md) - Methods - to_html() - [Flyte SDK > flyte.types > TypeEngine](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.types/typeengine.md) - Methods - dict_to_literal_map() - get_available_transformers() - get_transformer() - guess_python_type() - guess_python_types() - lazy_import_transformers() - literal_map_to_kwargs() - named_tuple_to_variable_map() - register() - register_additional_type() - register_restricted_type() - to_html() - to_literal() - to_literal_checks() - to_literal_type() - to_python_value() - unwrap_offloaded_literal() - [Flyte SDK > flyte.types > TypeTransformer](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.types/typetransformer.md) - Parameters - Properties - Methods - assert_type() - from_binary_idl() - get_literal_type() - guess_python_type() - isinstance_generic() - schema_match() - to_html() - to_literal() - to_python_value() - [Flyte SDK > flyte.types > TypeTransformerFailedError](https://www.union.ai/docs/v2/flyte/api-reference/flyte-sdk/flyte.types/typetransformerfailederror.md) - [Flyte agent plugins](https://www.union.ai/docs/v2/flyte/api-reference/agent-plugins.md) - Compatibility - Install Flyte Skills - ChatGPT - Install MCP Servers - Skills - SDK & workflow authoring - Migration (Flyte 1 → Flyte 2) - Migration (Slurm → Flyte 2) - Deployment - MCP servers - How skills and MCP servers work together - What each one does with your data - A cluster is optional - Adding MCP servers locally - Claude Code - ChatGPT desktop / Codex CLI - Hermes - OpenCode - Pi - Claude Code - ChatGPT desktop / Codex CLI - Hermes - OpenCode - Pi - Changing what is served - [Integrations](https://www.union.ai/docs/v2/flyte/api-reference/integrations.md) - [Integrations > Agent frameworks](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents.md) - [Integrations > Agent frameworks > Claude Agent SDK](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/claude.md) - Directory - Methods - Methods - [Integrations > Agent frameworks > Core](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/core.md) - Directory - Classes - Methods - Methods - [Integrations > Agent frameworks > Core > ReportTimeline](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/core/reporttimeline.md) - Parameters - Methods - heading() - row() - [Integrations > Agent frameworks > Core > ToolTaskResolver](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/core/tooltaskresolver.md) - Properties - Methods - load_app_env() - load_task() - loader_args() - [Integrations > Agent frameworks > CrewAI](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/crewai.md) - Directory - Methods - Methods - [Integrations > Agent frameworks > Deep Agents](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/deepagents.md) - Directory - Classes - Methods - Methods - [Integrations > Agent frameworks > Deep Agents > DurableChatModel](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/deepagents/durablechatmodel.md) - Parameters - Methods - bind_tools() - get_num_tokens() - get_num_tokens_from_messages() - [Integrations > Agent frameworks > Google ADK](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/google.md) - Directory - Classes - Methods - Methods - [Integrations > Agent frameworks > Google ADK > FlyteLlm](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/google/flytellm.md) - Parameters - Methods - generate_content_async() - [Integrations > Agent frameworks > Hermes](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/hermes.md) - Directory - Methods - Variables - Methods - [Integrations > Agent frameworks > LangChain](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/langchain.md) - Directory - Methods - Methods - [Integrations > Agent frameworks > LangGraph](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/langgraph.md) - Directory - Methods - Methods - [Integrations > Agent frameworks > Mistral](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/mistral.md) - Directory - Methods - Methods - [Integrations > Agent frameworks > OpenAI Agents SDK](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/openai.md) - Directory - Classes - Methods - Methods - [Integrations > Agent frameworks > OpenAI Agents SDK > FlyteModel](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/openai/flytemodel.md) - Parameters - Methods - close() - get_response() - stream_response() - [Integrations > Agent frameworks > OpenAI Agents SDK > FlyteModelProvider](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/openai/flytemodelprovider.md) - Parameters - Methods - aclose() - get_model() - [Integrations > Agent frameworks > OpenAI Agents SDK > FlyteSession](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/openai/flytesession.md) - Parameters - Methods - add_items() - clear_session() - get_items() - pop_item() - [Integrations > Agent frameworks > OpenAI Agents SDK > FlyteTracingProcessor](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/openai/flytetracingprocessor.md) - Parameters - Methods - force_flush() - on_span_end() - on_span_start() - on_trace_end() - on_trace_start() - shutdown() - [Integrations > Agent frameworks > OpenAI Agents SDK > FunctionTool](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/openai/functiontool.md) - Parameters - Methods - execute() - [Integrations > Agent frameworks > Pydantic AI](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/pydantic-ai.md) - Directory - Classes - Methods - Methods - [Integrations > Agent frameworks > Pydantic AI > FlyteModel](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agents/pydantic-ai/flytemodel.md) - Parameters - Properties - Methods - request() - request_stream() - [Integrations > Software development tools](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools.md) - [Integrations > Software development tools > ClickUp](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/clickup.md) - Directory - Classes - Methods - Variables - Methods - [Integrations > Software development tools > ClickUp > ClickUpProvider](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/clickup/clickupprovider.md) - Parameters - [Integrations > Software development tools > GitHub](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/github.md) - Human review gates - GitHub App tokens - Directory - Classes - Methods - Variables - Methods - [Integrations > Software development tools > GitHub > GitHubProvider](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/github/githubprovider.md) - Parameters - [Integrations > Software development tools > GitHub > ReviewComment](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/github/reviewcomment.md) - Parameters - [Integrations > Software development tools > GitHub > ReviewContext](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/github/reviewcontext.md) - Parameters - Methods - to_json() - [Integrations > Software development tools > GitHub > ReviewDecision](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/github/reviewdecision.md) - Parameters - Properties - [Integrations > Software development tools > Jira](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/jira.md) - Directory - Classes - Methods - Variables - Methods - [Integrations > Software development tools > Jira > JiraProvider](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/jira/jiraprovider.md) - Parameters - [Integrations > Software development tools > Linear](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/linear.md) - Directory - Classes - Methods - Variables - Methods - [Integrations > Software development tools > Linear > LinearProvider](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/linear/linearprovider.md) - Parameters - [Integrations > Software development tools > Slack](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/slack.md) - Directory - Classes - Methods - Variables - Methods - [Integrations > Software development tools > Slack > SlackProvider](https://www.union.ai/docs/v2/flyte/api-reference/integrations/software-development-tools/slack/slackprovider.md) - Parameters - [Integrations > Anthropic](https://www.union.ai/docs/v2/flyte/api-reference/integrations/anthropic.md) - Directory - Classes - Methods - Methods - [Integrations > Anthropic > Agent](https://www.union.ai/docs/v2/flyte/api-reference/integrations/anthropic/agent.md) - Parameters - Methods - get_anthropic_tools() - [Integrations > BigQuery](https://www.union.ai/docs/v2/flyte/api-reference/integrations/bigquery.md) - Directory - Classes - [Integrations > BigQuery > BigQueryConfig](https://www.union.ai/docs/v2/flyte/api-reference/integrations/bigquery/bigqueryconfig.md) - Parameters - [Integrations > BigQuery > BigQueryConnector](https://www.union.ai/docs/v2/flyte/api-reference/integrations/bigquery/bigqueryconnector.md) - Methods - create() - delete() - get() - get_logs() - get_metrics() - [Integrations > BigQuery > BigQueryTask](https://www.union.ai/docs/v2/flyte/api-reference/integrations/bigquery/bigquerytask.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Integrations > BigQuery > BQToPandasDecodingHandler](https://www.union.ai/docs/v2/flyte/api-reference/integrations/bigquery/bqtopandasdecodinghandler.md) - Parameters - Properties - Methods - decode() - [Integrations > Code generation](https://www.union.ai/docs/v2/flyte/api-reference/integrations/codegen.md) - Directory - Classes - [Integrations > Code generation > AutoCoderAgent](https://www.union.ai/docs/v2/flyte/api-reference/integrations/codegen/autocoderagent.md) - Parameters - Methods - generate() - [Integrations > Code generation > CodeGenEvalResult](https://www.union.ai/docs/v2/flyte/api-reference/integrations/codegen/codegenevalresult.md) - Parameters - Methods - as_task() - run() - [Integrations > Code generation > CodePlan](https://www.union.ai/docs/v2/flyte/api-reference/integrations/codegen/codeplan.md) - Parameters - [Integrations > Code generation > CodeSolution](https://www.union.ai/docs/v2/flyte/api-reference/integrations/codegen/codesolution.md) - Parameters - Methods - normalize_language() - [Integrations > Code generation > ErrorDiagnosis](https://www.union.ai/docs/v2/flyte/api-reference/integrations/codegen/errordiagnosis.md) - Parameters - [Integrations > Code generation > ImageConfig](https://www.union.ai/docs/v2/flyte/api-reference/integrations/codegen/imageconfig.md) - Parameters - [Integrations > Dask](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dask.md) - Directory - Classes - [Integrations > Dask > Dask](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dask/dask.md) - Parameters - [Integrations > Dask > Scheduler](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dask/scheduler.md) - Parameters - [Integrations > Dask > WorkerGroup](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dask/workergroup.md) - Parameters - [Integrations > Databricks](https://www.union.ai/docs/v2/flyte/api-reference/integrations/databricks.md) - Directory - Classes - [Integrations > Databricks > Databricks](https://www.union.ai/docs/v2/flyte/api-reference/integrations/databricks/databricks.md) - Parameters - [Integrations > Databricks > DatabricksConnector](https://www.union.ai/docs/v2/flyte/api-reference/integrations/databricks/databricksconnector.md) - Methods - create() - delete() - get() - get_logs() - get_metrics() - [Integrations > dbt](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dbt.md) - Directory - Classes - Errors - Methods - Methods - [Integrations > dbt > DbtInvocationError](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dbt/dbtinvocationerror.md) - Parameters - [Integrations > dbt > DbtNodeResult](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dbt/dbtnoderesult.md) - Parameters - [Integrations > dbt > DbtTask](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dbt/dbttask.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Integrations > dbt > DbtTaskResolver](https://www.union.ai/docs/v2/flyte/api-reference/integrations/dbt/dbttaskresolver.md) - Properties - Methods - load_task() - loader_args() - [Integrations > Gemini](https://www.union.ai/docs/v2/flyte/api-reference/integrations/gemini.md) - Directory - Classes - Methods - Methods - [Integrations > Gemini > Agent](https://www.union.ai/docs/v2/flyte/api-reference/integrations/gemini/agent.md) - Parameters - Methods - get_gemini_tools() - [Integrations > Grafana Agent Observability](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agento11y.md) - Directory - Classes - Methods - Variables - Methods - [Integrations > Grafana Agent Observability > FlyteIdentityBinding](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agento11y/flyteidentitybinding.md) - Parameters - Methods - step_span() - task_span() - [Integrations > Grafana Agent Observability > GrafanaAgentObservability](https://www.union.ai/docs/v2/flyte/api-reference/integrations/agento11y/grafanaagentobservability.md) - Parameters - Methods - get_link() - [Integrations > Hugging Face](https://www.union.ai/docs/v2/flyte/api-reference/integrations/huggingface.md) - Directory - Classes - Methods - Methods - [Integrations > Hugging Face > HFSource](https://www.union.ai/docs/v2/flyte/api-reference/integrations/huggingface/hfsource.md) - Parameters - Methods - from_hf_uri() - to_hf_uri() - [Integrations > Human-in-the-loop](https://www.union.ai/docs/v2/flyte/api-reference/integrations/hitl.md) - Basic usage: - Features: - Directory - Classes - Methods - Variables - Methods - [Integrations > Human-in-the-loop > Event](https://www.union.ai/docs/v2/flyte/api-reference/integrations/hitl/event.md) - Parameters - Properties - Methods - create() - wait() - [Integrations > Hydra](https://www.union.ai/docs/v2/flyte/api-reference/integrations/hydra.md) - Directory - Methods - Methods - [Integrations > JSONL](https://www.union.ai/docs/v2/flyte/api-reference/integrations/jsonl.md) - Directory - Classes - [Integrations > JSONL > JsonlDir](https://www.union.ai/docs/v2/flyte/api-reference/integrations/jsonl/jsonldir.md) - Parameters - Properties - Methods - download() - download_sync() - empty() - exists() - exists_sync() - from_existing_remote() - from_local() - from_local_sync() - get_file() - get_file_sync() - iter_arrow_batches() - iter_arrow_batches_sync() - iter_batches() - iter_batches_sync() - iter_records() - iter_records_sync() - list_files() - list_files_sync() - model_post_init() - new_remote() - walk() - walk_sync() - writer() - writer_sync() - [Integrations > JSONL > JsonlFile](https://www.union.ai/docs/v2/flyte/api-reference/integrations/jsonl/jsonlfile.md) - Parameters - Properties - Methods - download() - download_sync() - exists() - exists_sync() - from_existing_remote() - from_local() - from_local_sync() - iter_arrow_batches() - iter_arrow_batches_sync() - iter_records() - iter_records_sync() - model_post_init() - named_remote() - new_remote() - open() - open_sync() - writer() - writer_sync() - [Integrations > Lance](https://www.union.ai/docs/v2/flyte/api-reference/integrations/lance.md) - Directory - Methods - Methods - [Integrations > llama.cpp](https://www.union.ai/docs/v2/flyte/api-reference/integrations/llamacpp.md) - Directory - Classes - Methods - Variables - Methods - [Integrations > llama.cpp > LlamaCppAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/integrations/llamacpp/llamacppappenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - get_port() - on_shutdown() - on_startup() - server() - [Integrations > MLflow](https://www.union.ai/docs/v2/flyte/api-reference/integrations/mlflow.md) - Key features: - Basic usage: - Directory - Classes - Methods - Methods - [Integrations > MLflow > Mlflow](https://www.union.ai/docs/v2/flyte/api-reference/integrations/mlflow/mlflow.md) - Parameters - Methods - get_link() - [Integrations > NVIDIA Nsight Systems](https://www.union.ai/docs/v2/flyte/api-reference/integrations/nsight.md) - Directory - Functions - Packages - [Integrations > NVIDIA Nsight Systems > flyteplugins.nsight](https://www.union.ai/docs/v2/flyte/api-reference/integrations/nsight/flyteplugins.nsight.md) - Directory - Methods - Variables - Methods - [Integrations > NVIDIA Nsight Systems > flyteplugins.nsight.nsys](https://www.union.ai/docs/v2/flyte/api-reference/integrations/nsight/flyteplugins.nsight.nsys.md) - Directory - Methods - Methods - [Integrations > NVIDIA Nsight Systems > flyteplugins.nsight.nvtx](https://www.union.ai/docs/v2/flyte/api-reference/integrations/nsight/flyteplugins.nsight.nvtx.md) - Directory - Methods - Methods - [Integrations > OmegaConf](https://www.union.ai/docs/v2/flyte/api-reference/integrations/omegaconf.md) - Directory - Methods - Methods - [Integrations > OpenAI](https://www.union.ai/docs/v2/flyte/api-reference/integrations/openai.md) - Directory - Methods - Methods - [Integrations > OpenTelemetry](https://www.union.ai/docs/v2/flyte/api-reference/integrations/otel.md) - Directory - Classes - Methods - Methods - [Integrations > OpenTelemetry > OtelObserver](https://www.union.ai/docs/v2/flyte/api-reference/integrations/otel/otelobserver.md) - Parameters - Methods - step_span() - task_span() - [Integrations > OpenTelemetry > RunScopedIdGenerator](https://www.union.ai/docs/v2/flyte/api-reference/integrations/otel/runscopedidgenerator.md) - Parameters - Methods - generate_span_id() - generate_trace_id() - [Integrations > Pandera](https://www.union.ai/docs/v2/flyte/api-reference/integrations/pandera.md) - Directory - Classes - Methods - Methods - [Integrations > Pandera > ValidationConfig](https://www.union.ai/docs/v2/flyte/api-reference/integrations/pandera/validationconfig.md) - Parameters - [Integrations > Papermill](https://www.union.ai/docs/v2/flyte/api-reference/integrations/papermill.md) - Directory - Classes - Methods - Methods - [Integrations > Papermill > NotebookTask](https://www.union.ai/docs/v2/flyte/api-reference/integrations/papermill/notebooktask.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Integrations > Polars](https://www.union.ai/docs/v2/flyte/api-reference/integrations/polars.md) - Directory - Classes - Methods - Variables - Methods - [Integrations > Polars > ParquetToPolarsDecodingHandler](https://www.union.ai/docs/v2/flyte/api-reference/integrations/polars/parquettopolarsdecodinghandler.md) - Parameters - Properties - Methods - decode() - [Integrations > Polars > ParquetToPolarsLazyFrameDecodingHandler](https://www.union.ai/docs/v2/flyte/api-reference/integrations/polars/parquettopolarslazyframedecodinghandler.md) - Parameters - Properties - Methods - decode() - [Integrations > Polars > PolarsLazyFrameToParquetEncodingHandler](https://www.union.ai/docs/v2/flyte/api-reference/integrations/polars/polarslazyframetoparquetencodinghandler.md) - Parameters - Properties - Methods - encode() - [Integrations > Polars > PolarsToParquetEncodingHandler](https://www.union.ai/docs/v2/flyte/api-reference/integrations/polars/polarstoparquetencodinghandler.md) - Parameters - Properties - Methods - encode() - [Integrations > PyTorch](https://www.union.ai/docs/v2/flyte/api-reference/integrations/pytorch.md) - Directory - Classes - [Integrations > PyTorch > Elastic](https://www.union.ai/docs/v2/flyte/api-reference/integrations/pytorch/elastic.md) - Parameters - [Integrations > Ray](https://www.union.ai/docs/v2/flyte/api-reference/integrations/ray.md) - Directory - Classes - [Integrations > Ray > AutoscalerOptionsConfig](https://www.union.ai/docs/v2/flyte/api-reference/integrations/ray/autoscaleroptionsconfig.md) - Parameters - [Integrations > Ray > HeadNodeConfig](https://www.union.ai/docs/v2/flyte/api-reference/integrations/ray/headnodeconfig.md) - Parameters - [Integrations > Ray > RayJobConfig](https://www.union.ai/docs/v2/flyte/api-reference/integrations/ray/rayjobconfig.md) - Parameters - [Integrations > Ray > WorkerNodeConfig](https://www.union.ai/docs/v2/flyte/api-reference/integrations/ray/workernodeconfig.md) - Parameters - [Integrations > SGLang](https://www.union.ai/docs/v2/flyte/api-reference/integrations/sglang.md) - Directory - Classes - Variables - [Integrations > SGLang > SGLangAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/integrations/sglang/sglangappenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - get_port() - on_shutdown() - on_startup() - server() - [Integrations > Slurm](https://www.union.ai/docs/v2/flyte/api-reference/integrations/slurm.md) - Directory - Classes - [Integrations > Slurm > Slurm](https://www.union.ai/docs/v2/flyte/api-reference/integrations/slurm/slurm.md) - Parameters - Methods - to_custom_config() - [Integrations > Slurm > SlurmConnector](https://www.union.ai/docs/v2/flyte/api-reference/integrations/slurm/slurmconnector.md) - Parameters - Methods - create() - delete() - get() - get_logs() - get_metrics() - [Integrations > Slurm > SlurmFunctionTask](https://www.union.ai/docs/v2/flyte/api-reference/integrations/slurm/slurmfunctiontask.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Integrations > Slurm > SlurmJobMetadata](https://www.union.ai/docs/v2/flyte/api-reference/integrations/slurm/slurmjobmetadata.md) - Parameters - Methods - decode() - encode() - [Integrations > Slurm > SlurmScriptConnector](https://www.union.ai/docs/v2/flyte/api-reference/integrations/slurm/slurmscriptconnector.md) - Parameters - Methods - create() - delete() - get() - get_logs() - get_metrics() - [Integrations > Slurm > SlurmScriptTask](https://www.union.ai/docs/v2/flyte/api-reference/integrations/slurm/slurmscripttask.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Integrations > Snowflake](https://www.union.ai/docs/v2/flyte/api-reference/integrations/snowflake.md) - Directory - Classes - [Integrations > Snowflake > Snowflake](https://www.union.ai/docs/v2/flyte/api-reference/integrations/snowflake/snowflake.md) - Parameters - Properties - Methods - aio() - config() - container_args() - custom_config() - data_loading_config() - execute() - forward() - override() - post() - pre() - sql() - [Integrations > Snowflake > SnowflakeConfig](https://www.union.ai/docs/v2/flyte/api-reference/integrations/snowflake/snowflakeconfig.md) - Parameters - [Integrations > Snowflake > SnowflakeConnector](https://www.union.ai/docs/v2/flyte/api-reference/integrations/snowflake/snowflakeconnector.md) - Methods - create() - delete() - get() - get_logs() - get_metrics() - [Integrations > Spark](https://www.union.ai/docs/v2/flyte/api-reference/integrations/spark.md) - Directory - Classes - Methods - Methods - [Integrations > Spark > ParquetToSparkDecoder](https://www.union.ai/docs/v2/flyte/api-reference/integrations/spark/parquettosparkdecoder.md) - Parameters - Properties - Methods - decode() - [Integrations > Spark > Spark](https://www.union.ai/docs/v2/flyte/api-reference/integrations/spark/spark.md) - Parameters - [Integrations > Spark > SparkToParquetEncoder](https://www.union.ai/docs/v2/flyte/api-reference/integrations/spark/sparktoparquetencoder.md) - Parameters - Properties - Methods - encode() - [Integrations > Trackio](https://www.union.ai/docs/v2/flyte/api-reference/integrations/trackio.md) - Directory - Classes - Methods - Methods - [Integrations > Trackio > Trackio](https://www.union.ai/docs/v2/flyte/api-reference/integrations/trackio/trackio.md) - Parameters - Methods - get_link() - [Integrations > TypeSafe AI](https://www.union.ai/docs/v2/flyte/api-reference/integrations/typesafe-ai.md) - Directory - Classes - Errors - Methods - Variables - Methods - [Integrations > TypeSafe AI > BatteryError](https://www.union.ai/docs/v2/flyte/api-reference/integrations/typesafe-ai/batteryerror.md) - [Integrations > TypeSafe AI > CallInfo](https://www.union.ai/docs/v2/flyte/api-reference/integrations/typesafe-ai/callinfo.md) - Parameters - [Integrations > TypeSafe AI > Choice](https://www.union.ai/docs/v2/flyte/api-reference/integrations/typesafe-ai/choice.md) - Parameters - Methods - certain() - runner_up() - [Integrations > TypeSafe AI > MissingAPIKey](https://www.union.ai/docs/v2/flyte/api-reference/integrations/typesafe-ai/missingapikey.md) - [Integrations > TypeSafe AI > Noul](https://www.union.ai/docs/v2/flyte/api-reference/integrations/typesafe-ai/noul.md) - Parameters - Methods - at() - [Integrations > TypeSafe AI > Score](https://www.union.ai/docs/v2/flyte/api-reference/integrations/typesafe-ai/score.md) - Parameters - Methods - at_least() - [Integrations > vLLM](https://www.union.ai/docs/v2/flyte/api-reference/integrations/vllm.md) - Directory - Classes - Variables - [Integrations > vLLM > VLLMAppEnvironment](https://www.union.ai/docs/v2/flyte/api-reference/integrations/vllm/vllmappenvironment.md) - Parameters - Properties - Methods - add_dependency() - clone_with() - container_args() - container_cmd() - get_port() - on_shutdown() - on_startup() - server() - [Integrations > Weights & Biases](https://www.union.ai/docs/v2/flyte/api-reference/integrations/wandb.md) - Key features: - Basic usage: - Directory - Classes - Methods - Methods - [Integrations > Weights & Biases > Wandb](https://www.union.ai/docs/v2/flyte/api-reference/integrations/wandb/wandb.md) - Parameters - Methods - get_link() - [Integrations > Weights & Biases > WandbSweep](https://www.union.ai/docs/v2/flyte/api-reference/integrations/wandb/wandbsweep.md) - Parameters - Methods - get_link() - [LLM-optimized documentation](https://www.union.ai/docs/v2/flyte/api-reference/flyte-context.md) - Open a page in Claude or ChatGPT - One page as Markdown - An index of every page (`llms.txt`) - The entire site in one file (`llms-full.txt`) --- ## Community - [Joining the community](https://www.union.ai/docs/v2/flyte/community/joining-the-community.md) - Community sync - Contributor's sync - Newsletter - Slack guidelines - Abide by the [LF's Code of Conduct](https://lfprojects.org/policies/code-of-conduct/) - Avoid using DMs and @mentions - Make use of threads - Do not post the same question across multiple channels - Do not solicit members of our Slack - [Contributing code](https://www.union.ai/docs/v2/flyte/community/contributing-code.md) - Where the code lives - Filing an issue - Contributing to the Flyte SDK - Prerequisites - Set up your local environment - Set up prek pre-commit hooks - Make your changes - Sign off your commits (DCO) - Commit message guidelines - Open a pull request - Contributing with AI agent harnesses - Becoming a contributor - Resources - [Contributing docs and examples](https://www.union.ai/docs/v2/flyte/community/contributing-docs.md) - Where to start - How the docs site works - Versions - Common build infrastructure - Variants - Both Flyte and Union docs are open source - [Contributing docs and examples > Set up a local docs dev environment](https://www.union.ai/docs/v2/flyte/community/contributing-docs/quick-start.md) - Prerequisites - Clone the repository - Initialize the submodules - Configure the live preview - Start the live preview - Development settings - Build the production site - Next steps - [Contributing docs and examples > Authoring](https://www.union.ai/docs/v2/flyte/community/contributing-docs/authoring.md) - Getting started - Live preview - Page visibility - Page order - Page settings - Conditional content - Linking to the API reference - Sigils for special cases - Warnings and notices - Special content generation - Python generated content - Run on Union instructions - Jupyter notebooks - Mapped keys (`{{}}`) - Mermaid graphs - [Contributing docs and examples > Submit a contribution](https://www.union.ai/docs/v2/flyte/community/contributing-docs/submitting-contributions.md) - Target the right branch - Fork and clone - Create a feature branch - Make and preview your changes - Commit with a sign-off - Open a pull request - Check the preview build - Review and merge - When your change goes live - [Contributing docs and examples > Writing guidelines](https://www.union.ai/docs/v2/flyte/community/contributing-docs/writing-guidelines.md) - Content - Page structure - Voice - Sentences and words - Formatting - Code examples - Notices - Links - Images - Before you finish - [Contributing docs and examples > Variants](https://www.union.ai/docs/v2/flyte/community/contributing-docs/variants.md) - Variants at the whole-page level - Conditional rendering within a page - {{}} - {{}} - Full example - Adding a new variant - Location - Creating a new variant - Testing the new variant - Building (just) the variant - [Contributing docs and examples > Versions](https://www.union.ai/docs/v2/flyte/community/contributing-docs/versions.md) - What a version contains (and what it does not) - Lines are branches; versions are cuts - Which URL serves what - What this means when you contribute - Cutting a version - [Contributing docs and examples > Shortcodes](https://www.union.ai/docs/v2/flyte/community/contributing-docs/shortcodes.md) - How to specify a "shortcode" - Variants - Component library - `{{}}` - `{{}}` - `{{}}` - `{{}}` - `{{}}` - `{{}}` - `{{}}` and `{{}}` - `{{}}` - Deploy the {{%/* key product_name */%}} operator - Deploy the {{}} operator {#deploy-the-operator} - `{{}}` - `{{}}` - `{{}}`, `{{}}`, and `{{}}` - `{{}}` - `{{}}` - `{{}}` - `{{}}` and `{{}}` (legacy) - `{{}}` - [Contributing docs and examples > API docs](https://www.union.ai/docs/v2/flyte/community/contributing-docs/api-docs.md) - API naming convention - Package resource resolution - Tips and tricks - Auto-linking - Short vs. fully-qualified names - How auto-linking works - Sigils for inline code - [Contributing docs and examples > LLM-optimized documentation](https://www.union.ai/docs/v2/flyte/community/contributing-docs/llm-docs.md) - Output files - Discovery hierarchy - How the page twins are generated - The `llms-full.txt` link conversion - Regenerating - What readers see - [Contributing docs and examples > Redirects](https://www.union.ai/docs/v2/flyte/community/contributing-docs/redirects.md) - `docs.union.ai` redirects - `docs.flyte.org` redirects - [Contributing docs and examples > Production builds and troubleshooting](https://www.union.ai/docs/v2/flyte/community/contributing-docs/publishing.md) - Build the production site - Test the production build - Troubleshooting - Missing content - Page visibility --- ## Platform deployment - [Deployment overview](https://www.union.ai/docs/v2/flyte/oss-deployment/overview.md) - Architecture - Networking - External dependencies - Kubernetes cluster - Relational database - Object store - Optional dependencies - Ingress controller - DNS - SSL/TLS - Container images - [Kind deployment](https://www.union.ai/docs/v2/flyte/oss-deployment/kind-deployment.md) - 1. Prerequisites - Local machine - DigitalOcean VM - 2. Create the kind cluster - 3. Deploy the dependencies - PostgreSQL - Object store - AWS S3 - Cloudflare R2 - 4. Write the values file - 5. Install Flyte - 6. Access Flyte - Local machine - DigitalOcean VM - Submitting runs from the SDK - 7. Add authentication with a local IdP (optional) - Install the ingress controller - Deploy Dex - Deploy oauth2-proxy - Create the ForwardAuth middleware - Create the Flyte ingress with the middleware - Local machine - DigitalOcean VM - Verify the browser flow - Letting the SDK/CLI authenticate (with auth enabled) - Troubleshooting - 8. Load a local image into kind (optional) - 9. Tear down - [Kind deployment > Set up local OIDC provider](https://www.union.ai/docs/v2/flyte/oss-deployment/kind-deployment/local-oidc.md) - The issuer-URL problem - Dex configuration - Install Dex - Route the issuer path through Traefik - Next step - [Kind deployment > Set up external OIDC provider](https://www.union.ai/docs/v2/flyte/oss-deployment/kind-deployment/external-oidc.md) - Install the ingress controller - Deploy oauth2-proxy - Create the ForwardAuth middleware - Create the Flyte ingress with the middleware - Split the API and discovery paths off the browser middleware - Advertise your IdP to the SDK/CLI - Letting the SDK/CLI authenticate - Troubleshooting - [AWS deployment](https://www.union.ai/docs/v2/flyte/oss-deployment/aws-deployment.md) - [AWS deployment > flyte-binary deployment](https://www.union.ai/docs/v2/flyte/oss-deployment/aws-deployment/flyte-binary.md) - 1. Add the Helm repository - 2. Write a values file - 3. Install - 4. Grant object-store access - 5. Expose Flyte with an ingress - 6. Verify the installation - 7. Tear down - Full Values File Example - Default task resources - Default scheduling for task pods - OpenTelemetry - Configuring passwords from Secrets - [AWS deployment > flyte-core deployment](https://www.union.ai/docs/v2/flyte/oss-deployment/aws-deployment/flyte-core.md) - 1. Add the Helm repository - 2. Write a values file - 3. Install - 4. Grant object-store access - 5. Expose Flyte with an ingress - 6. Verify the installation - 7. Tear down - Full Values File Example - Scaling and scheduling individual components - Default task resources - Default scheduling for task pods - OpenTelemetry - Configuring passwords from Secrets - Values differences from flyte-binary - [Authentication and SSO](https://www.union.ai/docs/v2/flyte/oss-deployment/authentication.md) - Advertise your identity provider - Enforce auth at the ingress - Single sign-on for the console at the ALB - Prerequisites - 1. Configure the OIDC app - 2. Create the client Secret - 3. Allow the LB controller to read the Secret - 4. Add the auth annotations - 5. Verify - Troubleshooting - Annotation reference - Sign out - Cookies to expire - Troubleshooting - Run attribution (`executed_by`) - Behind a non-ALB proxy (oauth2-proxy / Traefik) - [Enable app serving](https://www.union.ai/docs/v2/flyte/oss-deployment/app-serving.md) - How it works - Prerequisites - 1. Install Knative Serving and Kourier - 2. Configure the apps domain - 3. Expose Kourier - LoadBalancer (simplest) - AWS (ALB in front) - 4. Enable app serving in Flyte - 5. Verify - Troubleshooting - [Plugin setup](https://www.union.ai/docs/v2/flyte/oss-deployment/plugin-setup.md) - [Plugin setup > Flyte copilot storage configuration](https://www.union.ai/docs/v2/flyte/oss-deployment/plugin-setup/copilot-storage.md) - The Secret the chart creates - S3 - Azure - GCS - Supplying your own Secret - With external configuration - S3 - Azure - GCS - Verify - Troubleshooting