TurboPuffer RAG Plugin
Hybrid search RAG (Retrieval Augmented Generation) implementation using TurboPuffer for vector + BM25 full-text search, with Gemini for embeddings.
Features
- Hybrid Search: Combines vector (semantic) and BM25 (keyword) search for better retrieval quality
- Reciprocal Rank Fusion: Merges results from multiple search strategies
- Gemini Embeddings: Uses Google's Gemini embedding model for high-quality vectors
- Low-latency Queries: Supports cache warming for fast query responses
- Implements RAG Interface: Compatible with Vision Agents RAG base class
Installation
uv add "vision-agents[turbopuffer]"
# or directly
uv add vision-agents-plugins-turbopuffer
Usage
from vision_agents.plugins import turbopuffer
# Initialize RAG
rag = turbopuffer.TurboPufferRAG(namespace="my-knowledge")
await rag.add_directory("./knowledge")
# Hybrid search (default)
results = await rag.search("How does the chat API work?")
# Vector-only search
results = await rag.search("How does the chat API work?", mode="vector")
# BM25-only search
results = await rag.search("chat API pricing", mode="bm25")
# Or use convenience function
rag = await turbopuffer.create_rag(
namespace="product-knowledge",
knowledge_dir="./knowledge"
)
Configuration
| Parameter | Description | Default |
|---|---|---|
namespace |
TurboPuffer namespace for storing vectors | Required |
embedding_model |
Gemini embedding model | models/gemini-embedding-001 |
chunk_size |
Size of text chunks for splitting documents | 10000 |
chunk_overlap |
Overlap between chunks for context continuity | 200 |
region |
TurboPuffer region | gcp-us-central1 |
Environment Variables
TURBO_PUFFER_KEY: TurboPuffer API keyGOOGLE_API_KEY: Google API key (for Gemini embeddings)
Dependencies
turbopuffer: TurboPuffer vector database clientlangchain-google-genai: Gemini embeddingslangchain-text-splitters: Text chunking utilities
References
2 maintainers
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file vision_agents_plugins_turbopuffer-0.6.5.tar.gz.
File metadata
- Download URL: vision_agents_plugins_turbopuffer-0.6.5.tar.gz
- Upload date:
- Size: 6.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.11.16 {"installer":{"name":"uv","version":"0.11.16","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
1ea2e576cac06c4abed6dd2a063ab4f691a9e5ab4be03d5a3c19ca912c324ed3
|
|
| MD5 |
a999e695c8c7d205af3773512502e848
|
|
| BLAKE2b-256 |
07d21a2cc1a4d2b88b9d6b6c8ff66727efbd69d55b0a38303d38c565b4c6948d
|
File details
Details for the file vision_agents_plugins_turbopuffer-0.6.5-py3-none-any.whl.
File metadata
- Download URL: vision_agents_plugins_turbopuffer-0.6.5-py3-none-any.whl
- Upload date:
- Size: 6.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.11.16 {"installer":{"name":"uv","version":"0.11.16","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
42fe34c2e0162b548730491802776f97c1475609db7c4e5abc71b1c7f08fed03
|
|
| MD5 |
8af4f71f958f184ad411f4c5657429b9
|
|
| BLAKE2b-256 |
06a382d3ca3402d1ffa8210b744846cd05d02cdb4a6a5f494528fdc05f25dd5e
|
