Reasoning on GPQA Diamond (Avg@8, Pass@8)
44.4Avg@8General-Reasoner-14B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| General-Reasoner-14BModel Category=Frontier Models2026.04 | 44.4 | 70.3 | |
| TTRLBase Model=Qwen3-8B-Base2026.04 | 41.1 | 73 | |
| EMPOBase Model=Qwen3-8B-Base2026.04 | 39.8 | 70.8 | |
| TEMPOBase Model=Qwen3-8B-Base2026.04 | 37.2 | 65.3 | |
| General-Reasoner-7BModel Category=Frontier Models2026.04 | 35.1 | 68.6 | |
| TEMPOBase Model=OLMO3-7B-Base2026.04 | 32.4 | 69.4 | |
| Zero-RL (PPO)Base Model=Qwen3-8B-Base2026.04 | 32.2 | 62.4 | |
| Olmo-3-7B-RL-Zero-GeneralModel Category=Frontier Models2026.04 | 28.9 | 69 | |
| TTRLBase Model=OLMO3-7B-Base2026.04 | 28.5 | 67.6 | |
| EMPOBase Model=OLMO3-7B-Base2026.04 | 27.7 | 61.6 | |
| Zero-RL (PPO)Base Model=OLMO3-7B-Base2026.04 | 21.9 | 62.1 | |
| MiMo-Zero-RL-7BModel Category=Frontier Models2026.04 | 18.8 | 45.8 |