# microsoft/X-Decoder

[CVPR 2023] Official Implementation of X-Decoder for generalized decoding for pixel, image and language

Repository: https://github.com/microsoft/X-Decoder
Canonical: https://ross.abutalabs.com/products/x-decoder
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2023-10-05T04:07:00+00:00

## Health v2 (maintenance only)
Score: 22/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 98
- inputs: {"age_days": 1375, "days_push": 1063, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1345, forks 161 (observed 2026-08-28T04:04:27.231238+00:00)

## What it is
Official PyTorch implementation of X-Decoder, a generalized decoding model from CVPR 2023 that unifies pixel-level segmentation, image-level understanding, and text-language tasks in one vision-language model. It includes demo, training, and evaluation code with pretrained checkpoints.

## Use cases
- segment objects in images with text prompts
- run referring expression segmentation
- caption images with a vision-language model
- evaluate panoptic and semantic segmentation models
- fine-tune a unified pixel-image-language decoder
- try a multimodal segmentation demo on HuggingFace

## When to choose
- you need a single model handling segmentation, retrieval, and captioning
- you want official pretrained checkpoints and evaluation benchmarks
- you are reproducing CVPR 2023 vision-language research

## When to avoid
- you need a lightweight production segmentation service
- you lack GPU hardware for inference or training
- you need actively maintained code with recent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, computer-vision, image-processing, nlp, deep-learning
- domain: computer-vision, machine-learning, image-processing, deep-learning
- platform: python
- tags: vision-language-model, segmentation, image-captioning, research-code, cvpr-2023, multimodal, natural-language-processing, linux, gpu

## Member repositories
- microsoft/X-Decoder (main) score 22

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:27.231238+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 2026-08-30T04:42:46.239634+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/X-Decoder (fetched 2026-08-28T04:04:27.231238+00:00, sha 70d6ee73165a)
- Data as of 2026-08-30T08:39:29.467469+00:00.