# facebookresearch/BLINK

Entity Linker solution

Repository: https://github.com/facebookresearch/BLINK
Canonical: https://ross.abutalabs.com/products/facebookresearch-blink
Language: Python
License: MIT
License Family: permissive
Archived: true
Last push: 2023-09-21T16:18:30+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2534, "days_push": 1077, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1210, forks 233 (observed 2026-08-28T04:03:59.996043+00:00)

## What it is
BLINK is a Python entity linking library from Facebook Research that resolves mentions in text to Wikipedia entities using a two-stage bi-encoder/cross-encoder BERT architecture. It includes FAISS-based dense retrieval and ELQ for end-to-end entity linking on questions.

## Use cases
- link entity mentions in text to wikipedia pages
- wikification of documents
- build a knowledge base entity linker
- dense retrieval of entities with faiss
- end-to-end entity linking for questions
- disambiguate mentions against a knowledge base

## When to choose
- you need state-of-the-art entity linking against Wikipedia
- you want dense bi-encoder retrieval with FAISS indexing
- you need end-to-end entity linking on questions via ELQ

## When to avoid
- you need entity linking against a custom non-Wikipedia knowledge base without adaptation
- you need a maintained library with recent updates or modern Python support
- you want a lightweight solution without downloading large pretrained models

## Facets
- artifact type: library
- maturity: maintenance
- function: nlp, machine-learning, search-engine, rag
- domain: artificial-intelligence, large-language-models
- platform: python
- tags: entity-linking, wikification, wikipedia, bert, faiss, bi-encoder, cross-encoder, facebook-research, natural-language-processing, search, linux, gpu

## Member repositories
- facebookresearch/BLINK (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:59.996043+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 2026-08-30T06:18:10.495101+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/BLINK (fetched 2026-08-28T04:03:59.996043+00:00, sha 8462e06d7a90)
- Data as of 2026-08-30T08:39:29.467469+00:00.