# KaiyangZhou/CoOp

Prompt Learning for Vision-Language Models (IJCV'22, CVPR'22)

Repository: https://github.com/KaiyangZhou/CoOp
Canonical: https://ross.abutalabs.com/products/coop
Language: Python
License: MIT
License Family: permissive
Topics: foundation-models, multimodal-learning, prompt-learning
Last push: 2024-05-20T16:58:40+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1827, "days_push": 835, "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 2221, forks 233 (observed 2026-08-28T04:06:27.610590+00:00)

## What it is
A research codebase implementing prompt learning methods (CoOp and CoCoOp) for adapting vision-language models like CLIP to downstream classification datasets. It accompanies CVPR 2022 and IJCV 2022 papers and provides training scripts, dataset loaders, and pre-trained weights.

## Use cases
- adapt CLIP to custom image classification datasets via prompt learning
- run few-shot image classification experiments with vision-language models
- reproduce CoOp and CoCoOp paper results
- learn context optimization for multimodal foundation models
- benchmark prompt tuning against linear probing
- fine-tune vision-language models without full model retraining

## When to choose
- you want to apply or extend CoOp/CoCoOp prompt learning on CLIP
- you need a reference implementation for few-shot vision-language adaptation research
- you want pre-trained CoOp weights for ImageNet-style classification

## When to avoid
- you need a production-ready, actively maintained ML library
- you want general-purpose model training unrelated to vision-language prompt tuning
- you need support for the latest prompt-learning methods beyond CoOp/CoCoOp

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, prompt-engineering, deep-learning
- domain: machine-learning, computer-vision, large-language-models
- platform: python
- tags: vision-language-models, clip, prompt-learning, few-shot-learning, transfer-learning, research-code, research, gpu

## Member repositories
- KaiyangZhou/CoOp (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:27.610590+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 2026-08-30T02:45:46.835456+00:00, confidence not recorded.
  - readme: https://github.com/KaiyangZhou/CoOp (fetched 2026-08-28T04:06:27.610590+00:00, sha 46835b585303)
- Data as of 2026-08-30T08:39:29.467469+00:00.