# IntelLabs/fastRAG

Efficient Retrieval Augmentation and Generation Framework

Repository: https://github.com/IntelLabs/fastRAG
Canonical: https://ross.abutalabs.com/products/fastrag
Language: Python
License: Apache-2.0
License Family: permissive
Topics: nlp, benchmark, colbert, information-retrieval, semantic-search, sentence-transformers, summarization, transformers, diffusion, knowledge-graph, multi-modal, question-answering, generative-ai, llm
Archived: true
Last push: 2026-01-12T19:58:22+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 62, release rhythm 40, longevity 94
- inputs: {"age_days": 1318, "days_push": 233, "days_rel": 646, "gap_med": 8.5, "n_releases_24m": 3}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1785, forks 166 (observed 2026-08-28T04:05:36.037227+00:00)

## What it is
fastRAG is a Python research framework for building efficient, optimized retrieval-augmented generation (RAG) pipelines, built on Haystack 2.0 and HuggingFace with components like ColBERT and optimized embedding models. It is specifically tuned for Intel hardware (Xeon CPUs and Gaudi accelerators) via IPEX, Optimum-Intel, and Optimum-Habana.

## Use cases
- build efficient rag pipelines
- run retrieval augmented generation on intel xeon or gaudi
- semantic search with colbert
- question answering over documents with llms
- benchmark rag components
- optimize embedding models with onnx runtime

## When to choose
- you need optimized RAG pipelines on Intel hardware
- you already use Haystack 2.0 and want efficient retriever/generator components
- you want ColBERT-based semantic search with index updating

## When to avoid
- you need actively maintained software - the project is archived with no further patches or releases
- you target non-Intel hardware where its optimizations provide no benefit
- you need production support or bug fixes from upstream

## Facets
- artifact type: framework
- maturity: abandoned
- function: rag, search-engine, llm-inference, nlp, machine-learning, benchmarking
- domain: large-language-models, machine-learning
- platform: python
- tags: haystack, colbert, intel-hardware, retrieval-augmented-generation, question-answering, semantic-search, archived, natural-language-processing, search, linux, gpu

## Member repositories
- IntelLabs/fastRAG (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:36.037227+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:24:11.335999+00:00, confidence not recorded.
  - readme: https://github.com/IntelLabs/fastRAG (fetched 2026-08-28T04:05:36.037227+00:00, sha 9c281a035b2f)
  - registry_pypi: https://pypi.org/pypi/fastrag/json (fetched 2026-08-29T11:02:30.356291+00:00, sha 1584b7b6916d)
- Data as of 2026-08-30T08:39:29.467469+00:00.