skills.rest
agtktID
agtktIDCOMMUNITY
Explore all skills by this author →
GitHub

qdrant-vector-search

Implement vector similarity search and RAG retrieval using the Qdrant database.

1|Updated Aug 6, 2026
Search Tags:#semantic search#embeddings#rag#vector database#vector search#qdrant#similarity search
One-click install
npx skills add https://github.com/agtktID/indagis-agent --skill qdrant-vector-search-agtktid
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: qdrant-vector-search
Source: https://github.com/agtktID/indagis-agent/tree/main/optional-skills/mlops/qdrant
Command: npx skills add https://github.com/agtktID/indagis-agent --skill qdrant-vector-search-agtktid

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires qdrant-client, and includes references (resource) components.

What problem does it solve? Building production RAG and semantic search systems requires a vector database that handles low-latency similarity search, metadata filtering, and scaling, which is complex to implement from scratch. ## Core Features & Use Cases - Vector Search with Filtering: Store dense and sparse vectors with payloads, then run nearest-neighbor queries combined with keyword, range, geo, and full-text filters. - RAG Integration: Connect with sentence-transformers, LangChain, and LlamaIndex to build retrieval pipelines for question answering. - Production Scaling: Use quantization, payload indexing, sharding, replication, and snapshots for large collections and distributed deployments. - Use Case: Index a knowledge base of documents as embeddings, then retrieve the top-k most relevant passages filtered by category to ground an LLM's answers. ## Quick Start Use the qdrant skill to create a collection, upsert document embeddings, and run a filtered similarity search against a local Qdrant instance.

Frequently Asked Questions about qdrant-vector-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform vector similarity search with Qdrant in Python?▼

Install qdrant-client, create a collection with VectorParams specifying dimension and distance metric, upsert PointStruct objects with vectors and payloads, then call client.query_points with your query vector and limit to retrieve nearest neighbors.

Qdrant vs Chroma vs FAISS for RAG applications?▼

Qdrant suits production RAG needing filtering, sharding, and on-premise control. Chroma is simpler for embedded use cases, FAISS offers maximum raw speed for batch research, and Pinecone fits fully managed zero-ops preferences.

Does Qdrant support hybrid search with sparse vectors?▼

Yes, Qdrant supports named dense and sparse vectors in one collection. You can combine both using prefetch queries with Reciprocal Rank Fusion to merge semantic and keyword search results.

Why is my Qdrant filtered search slow?▼

Slow filtered search usually means missing payload indexes on filtered fields. Create a payload index with client.create_payload_index for fields like category or timestamp, and consider enabling scalar quantization for large collections.

How do I fix vector dimension mismatch errors in Qdrant?▼

The error occurs when your embedding size differs from the collection's configured vector size. Check the expected dimension with client.get_collection, then recreate the collection with VectorParams matching your embedding model's output size.

When should I not use Qdrant for vector search?▼

Avoid Qdrant for simple embedded prototypes where Chroma's lighter setup suffices, or for pure batch research workloads where FAISS delivers maximum raw speed without server overhead.

Related Skills

Explore More Software Engineering Skills→
qcmuuqcmuu

sentence-transformers

Generate semantic embeddings from text using sentence-transformers models.

Community
Intermediate
collat-labscollat-labs
1

langchain-rag

Build retrieval augmented generation systems grounding LLM answers in external documents.

Official
Intermediate
PALabs-v1PALabs-v1
3

qdrant-hybrid-search-prefetches

Constructs prefetch queries for hybrid dense and sparse retrieval in Qdrant.

Official
Intermediate
MLT-OSSMLT-OSS
14

clip

Performs zero-shot image classification and image-text similarity using OpenAI's CLIP model.

Official
Intermediate