opendilab/LightRFT

Fine-tuning Tools

LightRFT: Light, Efficient, Omni-modal & Reward-model Driven Reinforcement Fine-Tuning Framework

GitHub →Website →

5.0

GitHub Metrics

Stars

397

Forks

Open Issues

Watchers

Contributors

Weekly Commits

Language

Python

License

Apache-2.0

Last Commit

Sep 7, 2026

Created

Dec 23, 2025

Latest Release

v0.1.1

Release Date

Jan 23, 2026

Synced: Oct 19, 2026

Quality Scores

Documentation Qualityw: 20%

5.7

Has docs site (https://opendilab.github.io/LightRFT). Description: 96 chars. Stars signal: 397. Contributors: 6. Score: 5.7/10

Community Healthw: 20%

3.1

Stars: 397. Contributors: 6. Watchers: 2. Forks: 12. Issue ratio: 3.5%. Score: 3.1/10

Maintenance Velocityw: 15%

5.4

Last commit: 42d ago. Weekly commits: 0. Latest release: v0.1.1. Score: 5.4/10

API Design & DXw: 20%

6.9

Stars/issues ratio: 28. Dynamic language: Python. Has documentation site. Permissive license: Apache-2.0. Popularity signal: 397 stars. Score: 6.9/10

Production Readinessw: 15%

3.5

Battle-tested: 397 stars. Peer review: 6 contributors. Versioned: v0.1.1. Licensed: Apache-2.0. Age: 0.8 years. Maintenance: last commit 42d ago. Score: 3.5/10

Ecosystem Integrationw: 10%

5.3

Fork interest: 12. Major ecosystem: Python. Integration-friendly: Apache-2.0. Adoption: 397 stars. Has web presence. Score: 5.3/10