Morphology Reasoning on Time-MMD Energy (test)
84.12AccuracyGPT-5.2
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5.2Horizon=96, Category=Advanced Models (reference only)2026.05 | 84.12 | |
| DeepSeek-R1Horizon=96, Category=Advanced Models (reference only)2026.05 | 79.51 | |
| KAIROSAGENT-4B (+ Turn-Level Reward RL)Horizon=96, Category=Comparable-Scale Models2026.05 | 50.47 | |
| KAIROSAGENT-4B (SFT-Only)Horizon=96, Category=Comparable-Scale Models2026.05 | 45.21 | |
| KAIROSAGENT-4B (+ Outcome-Level Reward RL)Horizon=96, Category=Comparable-Scale Models2026.05 | 43.33 | |
| Llama-3.1-8B-InstructHorizon=96, Category=Comparable-Scale Models2026.05 | 38.72 | |
| DeepSeek-R1-Distill-Qwen-7BHorizon=96, Category=Comparable-Scale Models2026.05 | 5.08 |