Knowledge Reasoning on GPQA Diamond (pass@1)
82.8pass@1Gemini-2.5 Flash-Thinking
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-2.5 Flash-ThinkingThinking Mode=true2025.12 | 82.8 | |
| DeepSeek-R1 0528 671BParameters=671B, Thinking Mode=true2025.12 | 81 | |
| Nemotron-Cascade 14B-ThinkingParameters=14B, Thinking Mode=true2025.12 | 69.6 | |
| Nemotron Cascade-8BParameters=8B, Thinking Mode=true2025.12 | 66.5 | |
| Nemotron-Nano 9B-v2Parameters=9B-v22025.12 | 64 | |
| Qwen3 14BParameters=14B2025.12 | 64 | |
| Qwen3 8BParameters=8B2025.12 | 62 |