Inferact/zero-mismatch-rl
Bitwise-identical token probabilities between RL rollout (vLLM) and training
GitHub repository with 40 stars and 5 forks.
24h trend summary
Trending score 0.61, freshness score 0.01, stars gained +0, forks gained +0.
Latest metric snapshot
2026-11-15: 40 stars and 5 forks.