THUDM /slime

slime is an LLM training framework for RL Scaling.

★ 8,810⑂ 1,355
PythonCreated 2025-06-18thudm.github.io/slime
8,810
stars
1,355
forks
–
watchers
–
prs
–
issues
–
releases
–
discussions

Snapshot: 2026-11-10T09:35:25Z · Data license Apache-2.0 · JSON