# facebookresearch/ImageBind

ImageBind One Embedding Space to Bind Them All

Repository: https://github.com/facebookresearch/ImageBind
Canonical: https://ross.abutalabs.com/products/imagebind
Language: Python
License: NOASSERTION
License Family: other
Last push: 2025-11-21T21:48:14+00:00

## Health v2 (maintenance only)
Score: 54/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 53, release rhythm 35, longevity 89
- inputs: {"age_days": 1259, "days_push": 285, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 9064, forks 843 (observed 2026-08-28T04:10:28.200643+00:00)

## What it is
A PyTorch library from Meta AI implementing ImageBind, a model that learns a joint embedding space across six modalities: images, text, audio, depth, thermal, and IMU data. It ships pretrained checkpoints and enables cross-modal retrieval, zero-shot classification, and modality arithmetic out of the box.

## Use cases
- embed images and text into a shared vector space
- cross-modal retrieval like finding images from audio or text
- zero-shot classification across modalities
- compose modalities with embedding arithmetic
- generate embeddings for depth, thermal, or IMU sensor data
- build multimodal search or RAG pipelines with unified embeddings

## When to avoid
- you only need standard image-text embeddings like CLIP with lighter dependencies
- you need production-grade inference serving rather than a research codebase
- you need modalities outside the six supported ones
- you require a permissive license - the license is custom, not standard open source

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, audio-processing, nlp
- domain: deep-learning, machine-learning, artificial-intelligence, computer-vision
- platform: python, windows
- tags: multimodal-embeddings, joint-embedding-space, cross-modal-retrieval, pytorch, pretrained-models, zero-shot-classification, research-model, audio, gpu, linux, macos

## Member repositories
- facebookresearch/ImageBind (main) score 54

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:28.200643+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 2026-08-29T17:24:01.845930+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/ImageBind (fetched 2026-08-28T04:10:28.200643+00:00, sha e0b74c87d636)
- Data as of 2026-08-30T08:39:29.467469+00:00.