初始化项目,由ModelHub XC社区提供模型

Model: devtaji/OpenThinker-Agent-repro-RL
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-01 09:43:18 +08:00
commit 96b140dc02
31 changed files with 304595 additions and 0 deletions

1
TRAINING_STATUS.md Normal file
View File

@@ -0,0 +1 @@
RL run rl-qwen3-8b-full728 (728 tasks, auto_snapshot=false). Checkpoints land here every 5 steps.