Fix: clamp rewards to (0.01, 0.99) for strict (0,1) range requirement e0aac36 helloAK96 commited on Apr 7
Major upgrade: composable rubrics, curriculum, multi-agent, analytics, 34 tasks 794e918 helloAK96 Claude Opus 4.6 (1M context) commited on Apr 5
Dynamic bug injection for all seeds + updated UI with sandbox mode 41a8917 helloAK96 Claude Opus 4.6 (1M context) commited on Apr 2
Add dynamic bug injection + expand to 23 tasks 63476bd helloAK96 Claude Opus 4.6 (1M context) commited on Mar 31
Initial commit: SQL Query Debugger OpenEnv environment 49ba367 helloAK96 Claude Opus 4.6 (1M context) commited on Mar 29