Public KG record
Agentic Reinforcement Learning
Recorded overview
RL stage for 8B/30B, learns to edit code, drive terminal, and run web searches i
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Evidence coverage
The KG currently records 6 known information dimensions and has 13 undocumented dimensions, linked to 3 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- Coding agents — uses
- OpenCode — uses
- GRPO — uses
- Codex — uses
- Granite 4.2 — trained on