bibirags 0.1.0

A simple, unified interface for RAG across multiple vector store backends (txtai, Chroma, Qdrant, LightRAG)

simplerags

A simple, unified interface for RAG (Retrieval-Augmented Generation) across multiple vector store backends.

simplerags wraps txtai, Chroma, Qdrant, and LightRAG behind a consistent three-function API so you can swap backends without rewriting your pipeline.

save_<backend>(chunks, rag_root, embed_model, ...)  → index documents
search_<backend>(query, rag_root, embed_model, ...) → retrieve chunks
query_<backend>(query, rag_root, llm_model, ...)    → retrieve + answer

All LLM and embedding calls go through LiteLLM, meaning any model provider (OpenAI, Anthropic, Cohere, Ollama, etc.) works out of the box.

---

Installation

Install the core package plus the backends you need:

# Qdrant only
pip install simplerags[qdrant]

# Chroma only
pip install simplerags[chroma]

# txtai only
pip install simplerags[txtai]

# LightRAG only
pip install simplerags[lightrag]

# All backends + document loading helpers
pip install simplerags[all]

Add [docs] to get PDF and TXT loading via LangChain:

pip install simplerags[qdrant,docs]

---

Quick start

Index raw text chunks

from simplerags import save_qdrant, search_qdrant, query_qdrant

chunks = [
    "The Eiffel Tower was completed in 1889.",
    "The Louvre is the world's largest art museum.",
    "Paris is the capital of France.",
]

embed_model = "text-embedding-3-small"   # any LiteLLM-compatible model
llm_model   = "gpt-4o-mini"
rag_root    = "./my_rag_index"

# 1. Index
save_qdrant(chunks, rag_root, embed_model)

# 2. Semantic search
results = search_qdrant("When was the Eiffel Tower built?", rag_root, embed_model)

# 3. RAG query → answer + source chunks
answer, sources = query_qdrant(
    "When was the Eiffel Tower built?", rag_root, llm_model, embed_model
)
print(answer)

Load documents from disk

from simplerags import chunk_docs, save_chroma, query_chroma

chunks = chunk_docs("./my_docs/", chunk_size=800, chunk_overlap=120)
save_chroma(chunks, "./chroma_index", embed_model="text-embedding-3-small")

answer, sources = query_chroma(
    "What does the contract say about termination?",
    rag_root="./chroma_index",
    llm_model="gpt-4o",
    embed_model="text-embedding-3-small",
)

Using Ollama (local models)

from simplerags import save_txtai, query_txtai

save_txtai(chunks, "./txtai_index", embed_model="ollama/bge-m3:latest")

answer, sources = query_txtai(
    "What happened in the news?",
    rag_root="./txtai_index",
    llm_model="ollama/gemma3:8b",
    embed_model="ollama/bge-m3:latest",
)

---

Backends at a glance

Backend	Best for	Index format	Notes
Qdrant	Production workloads, filtering	Local files or server	Cosine similarity, rich payload filtering
Chroma	LangChain ecosystems	Local SQLite	Easy LangChain integration
txtai	All-in-one HuggingFace pipelines	SQLite + FAISS	Built-in pipeline support
LightRAG	Knowledge-graph RAG	Local JSON + vector	Graph-enhanced hybrid retrieval

---

API reference

chunk_docs

chunk_docs(docs_path, chunk_size=800, chunk_overlap=120) → list[str]

Recursively loads .pdf and .txt files from docs_path and returns text chunks.

save_<backend>

save_qdrant(chunks, rag_root, embed_model)
save_chroma(chunks, rag_root, embed_model)
save_txtai(chunks, rag_root, embed_model)
save_lightrag(chunks, rag_root, embed_model, llm_model)

search_<backend>

results: list[str] = search_qdrant(query, rag_root, embed_model, top_k=3)

Returns the top_k most relevant chunk texts.

query_<backend>

answer, sources = query_qdrant(query, rag_root, llm_model, embed_model, top_k=3)

Returns (answer_string, list_of_source_chunks).

---

Contributing

git clone https://github.com/yourname/simplerags
cd simplerags
pip install -e ".[dev]"
pre-commit install
pytest

---

License

MIT – see LICENSE.