# openai/CLIP

CLIP (Contrastive Language-Image Pretraining),  Predict the most relevant text snippet given an image

Repository: https://github.com/openai/CLIP
Canonical: https://ross.abutalabs.com/products/clip
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: deep-learning, machine-learning
Last push: 2026-03-25T18:46:40+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 74, release rhythm 35, longevity 100
- inputs: {"age_days": 2086, "days_push": 161, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 34236, forks 4038 (observed 2026-08-28T04:11:58.761863+00:00)

## What it is
OpenAI's CLIP is a PyTorch library providing pretrained contrastive language-image models that encode images and text into a shared embedding space. It enables zero-shot image classification and cross-modal retrieval by predicting the most relevant text snippet for a given image.

## Use cases
- classify images into arbitrary categories with zero-shot learning
- compute similarity scores between images and text descriptions
- generate image and text embeddings for search
- build image search from natural language queries
- use CLIP embeddings as features for downstream models
- run zero-shot object recognition without labeled training data

## When to choose
- you need zero-shot image classification without training a custom model
- you want shared image-text embeddings for retrieval or search
- you need a well-tested PyTorch implementation of CLIP with pretrained checkpoints

## When to avoid
- you need fine-grained classification beyond CLIP's resolution
- you cannot run PyTorch or lack GPU/CPU compute for inference
- you need a lightweight model for edge devices with strict latency limits

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, computer-vision, nlp, image-processing
- domain: deep-learning, machine-learning, computer-vision, artificial-intelligence
- platform: python, cross-platform
- tags: clip, contrastive-learning, multimodal, zero-shot-classification, image-text-embedding, pytorch, vision-language-model, natural-language-processing, gpu

## Member repositories
- openai/CLIP (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:58.761863+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 2026-08-29T16:27:44.900715+00:00, confidence not recorded.
  - readme: https://github.com/openai/CLIP (fetched 2026-08-28T04:11:58.761863+00:00, sha a6b1c5372688)
- Data as of 2026-08-30T08:39:29.467469+00:00.