# facebookresearch/InferSent

InferSent sentence embeddings

Repository: https://github.com/facebookresearch/InferSent
Canonical: https://ross.abutalabs.com/products/infersent
Language: Jupyter Notebook
License: NOASSERTION
License Family: other
Archived: true
Last push: 2021-08-30T21:16:25+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": 3394, "days_push": 1829, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2280, forks 463 (observed 2026-08-28T04:06:33.789907+00:00)

## What it is
InferSent is a sentence embedding method from Facebook AI Research that produces semantic representations for English sentences using a model trained on natural language inference data. It provides pre-trained encoders (GloVe and fastText versions) plus a demo notebook for encoding sentences.

## Use cases
- generate sentence embeddings for English text
- encode sentences into semantic vectors for downstream tasks
- compute semantic similarity between sentences
- use pretrained NLI-trained sentence encoder in PyTorch
- get fixed-size vector representations of sentences for classification or retrieval

## When to choose
- you need classic pre-transformer sentence embeddings for English
- you want a simple PyTorch encoder with pretrained GloVe or fastText word vectors
- you are reproducing research from the InferSent paper or SentEval benchmarks

## When to avoid
- you need actively maintained software or support for languages other than English
- you want state-of-the-art contextual embeddings from transformer models
- you need multilingual sentence embeddings

## Facets
- artifact type: library
- maturity: abandoned
- function: nlp, machine-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: sentence-embeddings, pytorch, pretrained-models, natural-language-inference, research, natural-language-processing

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

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:33.789907+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 2026-08-30T02:41:22.500296+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/InferSent (fetched 2026-08-28T04:06:33.789907+00:00, sha 39af107821f3)
- Data as of 2026-08-30T08:39:29.467469+00:00.