# microsoft/i-Code

Repository: https://github.com/microsoft/i-Code
Canonical: https://ross.abutalabs.com/products/i-code
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2024-09-27T10:04:43+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 97
- inputs: {"age_days": 1365, "days_push": 705, "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 1703, forks 166 (observed 2026-08-28T04:05:24.474345+00:00)

## What it is
Microsoft's i-Code is a collection of research models and frameworks for integrative, composable multimodal AI spanning vision, language, and speech, including CoDi (any-to-any generation), UDOP (document intelligence), and i-Code Studio. It is primarily a research codebase accompanying published papers, implemented in Jupyter Notebook/Python.

## Use cases
- generate any output modality (image, video, audio, text) from any input modality
- build composable multimodal learning models combining vision, language, and speech
- unified document understanding and processing with vision, text, and layout
- autoregressive generation over vision, language, and speech data
- knowledge-based visual question answering
- reproduce research papers on multimodal foundation models

## When to choose
- you need state-of-the-art any-to-any multimodal generation like CoDi
- you want unified document AI with UDOP
- you are doing research on integrative multimodal learning and want reference implementations

## When to avoid
- you need a production-ready, well-supported library with stable APIs
- you want a simple pretrained model served via an API without GPU research setup
- you need lightweight inference on CPU or edge devices

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, nlp, speech-recognition, image-processing, llm-training
- domain: artificial-intelligence, machine-learning, computer-vision, speech-processing, deep-learning
- platform: python
- tags: multimodal, foundation-models, diffusion, document-intelligence, research, any-to-any-generation, composable-ai, natural-language-processing, gpu

## Member repositories
- microsoft/i-Code (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:24.474345+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 2026-08-30T03:37:38.381699+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/i-Code (fetched 2026-08-28T04:05:24.474345+00:00, sha 2b85ff7d36e9)
- Data as of 2026-08-30T08:39:29.467469+00:00.