# XueFuzhao/OpenMoE

A family of open-sourced Mixture-of-Experts (MoE) Large Language Models

Repository: https://github.com/XueFuzhao/OpenMoE
Canonical: https://ross.abutalabs.com/products/openmoe
Language: Python
License Family: other
Last push: 2024-03-08T15:08:26+00:00

## Health v2 (maintenance only)
Score: 28/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 80
- inputs: {"age_days": 1121, "days_push": 908, "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 1695, forks 86 (observed 2026-08-28T04:05:23.845162+00:00)

## What it is
OpenMoE is a family of open-sourced Mixture-of-Experts (MoE) large language models, including base and chat variants at 8B scale, released with training data, code, and model weights. It is a student-led research project aimed at growing the open-source MoE community and includes routing analysis research.

## Use cases
- run an open-source mixture-of-experts LLM
- study MoE routing behavior in large language models
- fine-tune or continue training an open MoE model
- convert JAX checkpoints to PyTorch
- research sparse expert models without closed weights

## When to choose
- you want fully open MoE weights, data, and training details for research
- you want to experiment with or extend an open MoE LLM
- you need a reproducible academic baseline for MoE LLM studies

## When to avoid
- you need a production-grade, actively maintained LLM with commercial support
- you require a permissive license for commercial use (no license is specified)
- you want the strongest possible model performance rather than research transparency

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: llm-training, machine-learning, deep-learning
- domain: large-language-models, machine-learning, artificial-intelligence
- platform: python, cloud
- tags: mixture-of-experts, open-source-models, jax, pytorch, research-project, model-weights, gpu

## Member repositories
- XueFuzhao/OpenMoE (main) score 28

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:23.845162+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 2026-08-30T03:37:55.871118+00:00, confidence not recorded.
  - readme: https://github.com/XueFuzhao/OpenMoE (fetched 2026-08-28T04:05:23.845162+00:00, sha 5b72b55f0665)
- Data as of 2026-08-30T08:39:29.467469+00:00.