Skip to content
J·BJon BishopFractional growth marketer

Data platforms · llms.txt file breakdown

Qdrant llms.txt

A directly retrieved Markdown index for Qdrant's documentation, with 708 links across 46 detected headings.

The actual file

Source Markdown

The saved llms.txt text, displayed as source. The file belongs to its publisher; the live endpoint may change.

Open saved text ↗

Showing the first 18000 characters. Load the complete file to read the remainder.

# https://qdrant.tech/
## Overall Summary
> Qdrant is an Open-Source Vector Search Engine written in Rust. It provides fast and scalable vector similarity search service with convenient API.

## Documentation
- [Documentation](https://qdrant.tech/documentation/index.md): Official Qdrant documentation for vector search and retrieval — quickstarts, deployment guides, integrations, and references for self-hosted and Qdrant Cloud.
- [Reranking for Better Search](https://qdrant.tech/documentation/search-precision/reranking-semantic-search/index.md): Improve semantic search and RAG quality in Qdrant by adding a reranking stage that re-scores top candidates with a cross-encoder for higher-precision results.
- [Qdrant Fundamentals](https://qdrant.tech/documentation/faq/qdrant-fundamentals/index.md): Common questions about Qdrant fundamentals — vector dimensions, payload limits, multivectors, real-time updates, and how similarity search behaves at scale.
- [Text Filtering](https://qdrant.tech/documentation/search/text-search/text-filtering/index.md): Apply keyword and text filters in Qdrant to narrow down search results by exact string values, multiple authors, or full-text terms — including progressive filtering via the batch search API.
- [Database Optimization](https://qdrant.tech/documentation/faq/database-optimization/index.md): FAQ on optimizing Qdrant: cut memory with quantization and on-disk storage, size machines correctly, and diagnose slow queries with payload indexing.
- [Full-Text Search](https://qdrant.tech/documentation/search/text-search/full-text-search/index.md): Use Qdrant's sparse vector support to perform full-text search with BM25, SPLADE++, and miniCOIL — including language-specific settings and parameter tuning.
- [Multivectors and Late Interaction](https://qdrant.tech/documentation/tutorials-search-engineering/using-multivector-representations/index.md): Tutorial: use Qdrant multivectors with late interaction reranking to capture token-level signals and improve precision while controlling RAM and indexing cost.
- [Relevance Feedback Retrieval in Qdrant](https://qdrant.tech/documentation/tutorials-search-engineering/using-relevance-feedback/index.md): Tutorial: integrate Qdrant's relevance feedback query into a search pipeline, customize feedback parameters, and evaluate retrieval quality gains.
- [Collaborative Filtering](https://qdrant.tech/documentation/tutorials-search-engineering/collaborative-filtering/index.md): Build an effective movie recommendation system using collaborative filtering and Qdrant's similarity search.
- [Hugging Face Dataset Ingestion](https://qdrant.tech/documentation/tutorials-basics/huggingface-datasets/index.md): Step-by-step guide to ingest Hugging Face datasets into Qdrant collections using the Python SDK, with batching and streaming for large vector workloads.
- [Hybrid Search](https://qdrant.tech/documentation/search/text-search/hybrid-search/index.md): Build hybrid search pipelines in Qdrant by combining dense semantic vectors and sparse lexical vectors with the prefetch API and Reciprocal Rank Fusion for comprehensive search results.
- [Async API](https://qdrant.tech/documentation/tutorials-develop/async-api/index.md): Tutorial: use Qdrant's async Python API with FastAPI and other ASGI frameworks to build non-blocking, high-throughput vector search applications.
- [Multivector Document Retrieval](https://qdrant.tech/documentation/tutorials-search-engineering/pdf-retrieval-at-scale/index.md): Optimizing PDF retrieval at scale with Qdrant and Vision Large Language Models (VLLMs) such as ColPali and ColQwen. Two-stage retrieval with multivector representations mean pooling.
- [GraphRAG with Qdrant and Neo4j](https://qdrant.tech/documentation/examples/graphrag-qdrant-neo4j/index.md): A step-by-step guide on implementing GraphRAG with Qdrant and Neo4j, combining vector search with knowledge graphs for enhanced retrieval.
- [Multimodal Food Search Demo](https://qdrant.tech/documentation/tutorials-search-engineering/food-discovery-recommendations/index.md): Tutorial: explore the Food Discovery demo, which uses CLIP embeddings, grouped search, recommendations with best_score and average_vector strategies, and geo filters in Qdrant.
- [Agentic RAG with CrewAI](https://qdrant.tech/documentation/tutorials-build-essentials/agentic-rag-crewai-zoom/index.md): Step-by-step tutorial: build an agentic RAG system with Qdrant and CrewAI to store, retrieve, and analyze meeting transcripts through coordinated AI agents.
- [Configuration](https://qdrant.tech/documentation/ops-configuration/configuration/index.md): Configure Qdrant via config files and environment variables to set storage paths, network interfaces, performance parameters, and feature flags.
- [Create a Cluster](https://qdrant.tech/documentation/cloud/create-cluster/index.md): Create a free or standard Qdrant Cloud cluster on AWS, GCP, or Azure, with high availability, GPU acceleration, and Hybrid Cloud deployment options.
- [Databricks Ingestion](https://qdrant.tech/documentation/send-data/databricks/index.md): Use FastEmbed on Databricks to generate dense and sparse vectors, upload a 100-row dataset to Qdrant, and verify the ingestion.
- [Edge vs. Qdrant Cluster](https://qdrant.tech/documentation/edge/edge-vs-qdrant-cluster/index.md): Compare Qdrant Edge with Qdrant Server across architecture, deployment, operations, and API surface to decide which fits your use case.
- [From Pinecone](https://qdrant.tech/documentation/migrate-to-qdrant/from-pinecone/index.md): Migrate from Pinecone to Qdrant by streaming serverless index data, vectors, namespaces, and metadata into Qdrant collections with the Qdrant Migration Tool.
- [Keeping Postgres and Qdrant in Sync](https://qdrant.tech/documentation/data-synchronization/with-postgres/index.md): Sync Postgres with Qdrant using dual-write, queued, or Change Data Capture patterns to keep vector search aligned with your relational source of truth.
- [Measuring ANN Recall](https://qdrant.tech/documentation/tutorials-search-engineering/ann-recall/index.md): Measuring ANN Recall Time: 15 min Level: Beginner This tutorial focuses on ANN recall: how closely approximate nearest-neighbor (ANN) search matches exact kNN search, measured with recall@k.
- [Monitoring & Telemetry](https://qdrant.tech/documentation/ops-monitoring/monitoring/index.md): Scrape Qdrant metrics in Prometheus and OpenMetrics format and visualize cluster, collection, and node-level telemetry in Grafana monitoring dashboards.
- [OpenLLMetry](https://qdrant.tech/documentation/observability/openllmetry/index.md): Instrument the Qdrant client with OpenLLMetry, an OpenTelemetry-based observability layer from Traceloop, and export traces to any OTel-compatible backend.
- [Optimize Performance](https://qdrant.tech/documentation/ops-optimization/optimize/index.md): Optimize Qdrant performance with quantization, on-disk vector storage, and HNSW tuning to balance search speed, precision, and memory consumption.
- [Points](https://qdrant.tech/documentation/manage-data/points/index.md): Manage points in Qdrant collections by upserting, updating, deleting, and retrieving records that combine a vector with an optional JSON payload for search.
- [Pre-Migration Baseline](https://qdrant.tech/documentation/migration-guidance/pre-migration-baseline/index.md): Record a pre-migration baseline of collection inventory, metadata schema, and reference search results so you can verify a clean migration cutover to Qdrant.
- [Pulumi](https://qdrant.tech/documentation/cloud-tools/pulumi/index.md): Use Pulumi to provision and manage Qdrant Cloud clusters as code, defining clusters, API keys, and packages in Python, TypeScript, Go, or other supported languages.
- [Role Management](https://qdrant.tech/documentation/cloud-rbac/role-management/index.md): Create, edit, and assign custom Qdrant Cloud roles with fine-grained permissions, or use built-in Base, Admin, and Owner roles for access control.
- [Search](https://qdrant.tech/documentation/search/search/index.md): Run similarity search in Qdrant to retrieve nearest-neighbor points for text, image, and multimodal queries using configurable distance metrics and HNSW.
- [Setup Hybrid Cloud](https://qdrant.tech/documentation/hybrid-cloud/hybrid-cloud-setup/index.md): Onboard a Kubernetes cluster as a Qdrant Hybrid Cloud Environment by installing the Operator and Cloud Agent with the required storage and permissions.
- [Setup Private Cloud](https://qdrant.tech/documentation/private-cloud/private-cloud-setup/index.md): Install Qdrant Private Cloud into any standard Kubernetes cluster with Helm, the Qdrant Operator, and credentials for the Qdrant container registry.
- [Vertical Scaling](https://qdrant.tech/documentation/scaling/vertical-scaling/index.md): Learn when and how to scale Qdrant vertically by resizing node resources, including RAM sizing formulas, Qdrant Cloud and self-hosted resize steps, and the signals that mean it's time to scale horizontally instead.
- [What is Qdrant?](https://qdrant.tech/documentation/overview/what-is-qdrant/index.md): Qdrant is a vector database for storing, searching, and managing high-dimensional embeddings — purpose-built for semantic search, RAG, and recommendations.
- [Measuring Retrieval Relevance](https://qdrant.tech/documentation/search-evaluation/retrieval-relevance/index.md): Measure Qdrant retrieval relevance with labeled queries, document IDs, and ranking metrics to compare search configurations on your own data.
- [Evaluating Pipeline Output Quality](https://qdrant.tech/documentation/search-evaluation/pipeline-output-quality/index.md): Evaluate retrieval and generation separately to identify why a Qdrant search pipeline returns an unsupported answer or misses the information users need.
- [Memory Usage](https://qdrant.tech/documentation/ops-monitoring/memory-usage/index.md): Use the Qdrant Web UI or API to monitor a collection's disk, RAM, and page cache consumption across the cluster and per component, to plan capacity and diagnose memory pressure.
- [Resource Quotas](https://qdrant.tech/documentation/ops-configuration/quotas/index.md): Configure cluster-wide memory and disk quotas in Qdrant, understand HTTP 507 rejections and replica deactivation, and find which node in a cluster is full.
- [Memory Tiers in Qdrant: What to Use and When](https://qdrant.tech/documentation/production-operations/memory-tiers/index.md): Which Qdrant memory tier layout to use and when, and why, backed by benchmarks.
- [Slow Request Log](https://qdrant.tech/documentation/ops-monitoring/slow-request-log/index.md): Use the Slow Request Log REST endpoint to find which queries are responsible for high latency, with per-request-type breakdowns, deduplication by request content, and approximate occurrence counts.
- [How to Choose an Embedding Model: Evaluation & Tradeoffs](https://qdrant.tech/documentation/search-patterns/choose-embedding-model/index.md): Building proper search requires selecting the right embedding model for your specific use case. This guide helps you navigate the selection process based on performance, cost, and other practical considerations.
- [Administration](https://qdrant.tech/documentation/ops-configuration/administration/index.md): Manage Qdrant at runtime with administration tools for recovery mode, strict mode, and guardrails against expensive queries on unindexed payload fields.
- [Agent Skills](https://qdrant.tech/documentation/agentic-tools/skills/index.md): Qdrant agent skills help AI coding assistants diagnose and tune vector search in production. Pass a skill URL from skills.qdrant.tech to your agent to get targeted guidance on scaling, search quality, performance, monitoring, and more.
- [Authentication](https://qdrant.tech/documentation/cloud/authentication/index.md): Create Qdrant Cloud Database API keys with granular access control, configure expirations, and authenticate to your clusters with best-practice rotation.
- [Configuration](https://qdrant.tech/documentation/private-cloud/configuration/index.md): Configure the Qdrant Private Cloud Helm chart with Operator settings, networking, scheduling, ingress, and backup management options for Kubernetes.
- [Create a Cluster](https://qdrant.tech/documentation/hybrid-cloud/hybrid-cloud-cluster-creation/index.md): Create a production-ready Qdrant cluster in Hybrid Cloud with Kubernetes scheduling controls, node selectors, and secret-based API key authentication.
- [Data Integrity](https://qdrant.tech/documentation/migration-guidance/data-integrity/index.md): Run data integrity checks after migrating to Qdrant by comparing vector counts, sample records, and payload fields against your source vector database.
- [Filtering](https://qdrant.tech/documentation/search/filtering/index.md): Filter Qdrant search results with payload conditions on metadata and IDs, combining database-style clauses with vector similarity for precise retrieval.
- [From Weaviate](https://qdrant.tech/documentation/migrate-to-qdrant/from-weaviate/index.md): Migrate from Weaviate to Qdrant by streaming classes, vectors, and properties into Qdrant collections using the Qdrant Migration Tool's Weaviate connector.
- [Horizontal Scaling](https://qdrant.tech/documentation/scaling/horizontal-scaling/index.md): Learn how Qdrant's distributed model works: sharding, the default replication behavior, Raft consensus for cluster metadata, and the consistency trade-off.
- [Inference API](https://qdrant.tech/documentation/inference/inference-api/index.md): Inference API When using Qdrant Cloud Inference or Qdrant’s BM25 model, embeddings are generated server-side. Instead of pre-computed vectors, you pass Inference Objects when ingesting or
- [LLM-Powered Filter Automation](https://qdrant.tech/documentation/search-precision/automate-filtering-with-llms/index.md): Automate filter construction in Qdrant by using an LLM to convert natural-language queries into structured payload filter conditions for precise vector search.
- [OpenLIT](https://qdrant.tech/documentation/observability/openlit/index.md): Use OpenLIT, an OpenTelemetry-native LLM observability tool, to auto-instrument Qdrant and forward traces and metrics to your existing telemetry backend.
- [Optimizer](https://qdrant.tech/documentation/ops-optimization/optimizer/index.md): See how Qdrant's background optimizer reclaims deleted points, merges segments, and rebuilds indexes to keep storage compact and search performance high.
- [Qdrant Cloud Inference](https://qdrant.tech/documentation/embeddings/qdrant-cloud-inference/index.md): Use Qdrant Cloud Inference to embed text and images server-side: hosted dense, sparse, and multivector models, plus OpenAI, Cohere, Jina AI, and OpenRouter.
- [Querying with Airflow](https://qdrant.tech/documentation/send-data/qdrant-airflow-astronomer/index.md): Build Apache Airflow DAGs on Astronomer that generate embeddings in parallel and run semantic queries against Qdrant for production data pipelines.
- [Quickstart](https://qdrant.tech/documentation/edge/edge-quickstart/index.md): Set up Qdrant Edge with Python or Rust bindings to configure shards, upsert points, build payload indexes, and run local vector search on-device.
- [Quickstart](https://qdrant.tech/documentation/fastembed/fastembed-quickstart/index.md): Install FastEmbed, load a default text embedding model, and generate vectors from sample documents in this quickstart for Qdrant's lightweight embedding library.
- [S3 Ingestion](https://qdrant.tech/documentation/tutorials-build-essentials/data-ingestion-beginners/index.md): Tutorial: build a data ingestion pipeline in plain Python that reads reviews and images from S3 or S3-compatible storage, embeds them with FastEmbed, and stores them in Qdrant for semantic search.
- [Scaling and Monitoring Hybrid Cloud](https://qdrant.tech/documentation/tutorials-operations/scaling-and-monitoring-pods/index.md): Scaling and Monitoring Hybrid Cloud Time: 25 min Level: Intermediate Stack: Kubernetes, Qdrant Hybrid Cloud Scaling a Qdrant cluster on Hybrid Cloud looks like a Kubernetes problem from the outside:
- [Semantic Search 101](https://qdrant.tech/documentation/tutorials-basics/search-beginners/index.md): Step-by-step tutorial: spin up a Qdrant Cloud cluster, create a collection, upload sample data, and run semantic vector search queries in five minutes.
- [Shard Lifecycle](https://qdrant.tech/documentation/edge/edge-api/shard-lifecycle/index.md): Reference for the Qdrant Edge Shard lifecycle methods: creating a new shard, loading an existing one, inspecting its path and configuration, reading shard metadata with info, flushing to disk, and closing it.
- [Snapshots](https://qdrant.tech/documentation/tutorials-operations/create-snapshot/index.md): Tutorial: create, download, and restore Qdrant snapshots to back up collections or migrate data between single-node and distributed multi-node clusters.
- [Terraform](https://qdrant.tech/documentation/cloud-tools/terraform/index.md): Provision Qdrant Cloud clusters and database keys with the official Terraform provider to manage vector database infrastructure as code on AWS, GCP, and Azure.
- [Understanding Vector Search in Qdrant](https://qdrant.tech/documentation/overview/vector-search/index.md): Learn how vector search uses embeddings to retrieve by semantic meaning, how it differs from keyword search, and where Qdrant fits in the retrieval stack.
- [User Management](https://qdrant.tech/documentation/cloud-rbac/user-management/index.md): Invite teammates, assign or revoke roles, transfer account ownership, and remove users from your Qdrant Cloud account through User Management.
- [Vectors](https://qdrant.tech/documentation/manage-data/vectors/index.md): Define dense, sparse, and multivector configurations in Qdrant collections to support text, image, multimodal, and late-interaction retrieval like ColBERT.
- [Multi-Representation Search](https:/

Measured from the saved file

Linked document

UTF-8 text size
180.6 KB184981 bytes
Token count
41636o200k_base encoding
Markdown links
706706 distinct URLs
Fenced code blocks
0797 lines in the file

Size measures the captured UTF-8 text, not compressed transfer size. Tokens are counted with a named tokenizer; other models can count differently. Neither measure establishes context compatibility or answer quality.

Project H1
https://qdrant.tech/

Context architecture

How the content is divided

45 H2 sections organize 706 parsed Markdown links. 706 links point to .md or .markdown paths; 0 link occurrences use another hostname. Linked destinations have not been recursively validated.

H2 sectionLinksUTF-8 bytesTokens
Overall SummaryLine 2015029
DocumentationLine 53087856217533
Qdrant AcademyLine 315112288046436
Advanced Data Analysis with Anomaly Detection & Entity MatchingLine 429125048
Advanced Search Solutions: High-Performance Vector SearchLine 432125949
AgendaLine 435124567
AI Agents with QdrantLine 438128561
brand-resourcesLine 44118423
Building production vector search shouldn’t be so hard.Line 444123554
CommunityLine 447117140
Contact QdrantLine 450115740
CustomersLine 45316617
Enterprise Search Solutions for Your Business | QdrantLine 456127356
EventsLine 45915716
GPU Indexing | Build Vector Indexes Up to 4x Faster on Qdrant CloudLine 462133888
Healthcare AI & MedTechLine 465124450
Show 29 more sections
H2 sectionLinksUTF-8 bytesTokens
LegalLine 46851259294
PartnersLine 47516317
Pricing for Cloud and Vector Database Solutions QdrantLine 478125750
private-cloudLine 48117819
Qdrant BlogLine 4841985471012577
Qdrant Cloud Inference | Native Embeddings in your Vector Database for Multimodal and Hybrid SearchLine 684145495
Qdrant Cloud Observability: Metrics, Monitoring, and AlertingLine 687128468
Qdrant Cloud Resilience: Replication, Failover, and Multi-AZLine 690127371
Qdrant Cloud: Scalable Managed Cloud ServicesLine 693124252
Qdrant DemosLine 696126450
Qdrant Edge | Embedded and Edge AI SystemsLine 699127461
Qdrant for StartupsLine 702123158
Qdrant High-Performance Vector Search EngineLine 705125752
Qdrant Hybrid Cloud: Flexible Deployment, Data Privacy, and Cost EfficiencyLine 708133669
Qdrant Learn PortalLine 7112453100
Qdrant Quantization: Cut Vector Memory, Keep RecallLine 715126259
Qdrant Security: Compliance, Isolation, and Encryption Across Every DeploymentLine 7182744149
Qdrant ServerlessLine 722125359
Qdrant Services | Implementation Partners for Vector Search and AILine 725129362
Qdrant StarsLine 72819327
RAG Use Case: Advanced Vector Search for AI ApplicationsLine 731234780
Recommendation Engines: Personalization & Data HandlingLine 735121839
SubscribeLine 73817919
SubscribeLine 74116617
Technical ArticlesLine 7443690521971
Vector Search BenchmarksLine 78261779387
Vector Search Use CasesLine 790121845
Vector Space DayLine 79319326
Vector Space Day SF 2026 RecapLine 796119151

Link counts exclude parsed fenced and inline code and resolve relative destinations against the final endpoint. Section sizes cover the text below each H2 until the next H2.

Embedded material

Code, installation and API examples

Fenced blocks
0 closed
Declared languages
No fenced blocks detected
Package-install commands
0 detected in fenced or inline code
Authentication examples
0 fenced lines match API-key, Authorization, Bearer, OAuth or access-token patterns
HTTP request examples
0 fenced lines match request-method patterns
Authentication links
0 links have authentication-related labels

Embedded example text and links to further documentation are counted separately. A code block does not establish that a sample runs or that an authentication flow succeeds.

Publisher prioritization

The Optional section

No H2 section named “Optional” was detected in this file.

The community proposal uses Optional by convention for secondary links that can be skipped for a shorter context. This label does not establish that everything outside it is critical, or that the material inside it is deprecated.

Machine-readable routes

MCP references and configuration text

MCP-related text was detected in the captured file. 2 links have MCP-related labels or paths. 0 fenced lines match mcpServers, mcp_servers or an mcp add command.

These are text and link observations. No MCP server connection, authentication or tool call was tested.

Index and bundle

llms-full.txt companion

Not found at checked URLs

A verified text response establishes a companion at the checked URL. It does not prove the bundle contains all product documentation, that generation is automated, or that it fits an assistant’s context window.

Change evidence

What can be said about freshness

This capture establishes a baseline. Update frequency cannot be determined from one observation.

Linked release and migration material

10 links have release, changelog, deprecation or migration terms in their labels or paths. These provide places to investigate product changes separately.

File changes can reveal documentation edits or navigation changes. They do not by themselves measure product release velocity, accuracy, agent adoption or AI visibility.

Capture identity and measurement method
Requested endpoint
https://qdrant.tech/llms.txt
Final endpoint
https://qdrant.tech/llms.txt
HTTP status
200
Content type
text/plain; charset=UTF-8
Text retrieval
Firecrawl literal text capture
Tokenizer
tiktoken 0.14.0 · o200k_base
SHA-256 of saved UTF-8 text
25db2947aa70017665921aee47598c542ff6887ab54a7256717cf065e6b11229

Counts use a deterministic parser, not a spec-compliance grade. Unusual Markdown and indented code can fall outside its recognized patterns. See the llms.txt proposal and tokenizer implementation.

Related tools and resources

Contact Jon