Evolution on OneReason-Bench
58.92Selection ScoreGPT-5.4
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-5.4Category=Text-Based2026.06 | 58.92 | 41.41 | 17.61 | |
| Deepseek-V3.2Category=Text-Based2026.06 | 57.18 | 27.13 | 11.32 | |
| Claude-Opus-4.6Category=Text-Based2026.06 | 56.84 | 17.16 | 13.46 | |
| Gemini-3-PreviewCategory=Text-Based2026.06 | 56.83 | 33.68 | 14.76 | |
| Qwen3-32BCategory=Text-Based2026.06 | 51.96 | 28.05 | 7.73 | |
| OneReason RFT thinkingCategory=Itemic Token-Based, Thinking Mode=thinking, Stage=RFT2026.06 | 42.42 | 39.57 | 21.23 | |
| Qwen3-8BCategory=Text-Based2026.06 | 40.7 | 25.49 | 8.6 | |
| OneReason RFT non-thinkingCategory=Itemic Token-Based, Thinking Mode=non-thinking, Stage=RFT2026.06 | 38.92 | 39.33 | 20.31 | |
| OneReason SFT non-thinkingCategory=Itemic Token-Based, Thinking Mode=non-thinking, Stage=SFT2026.06 | 35.07 | 33.87 | 15.42 | |
| OneReason SFT thinkingCategory=Itemic Token-Based, Thinking Mode=thinking, Stage=SFT2026.06 | 32.18 | 31.6 | 14.31 | |
| LC-Rec-PT-SFT-8BCategory=Itemic Token-Based2026.06 | 3.32 | 8.6 | 4.46 | |
| LC-Rec-SFT-Only-8BCategory=Itemic Token-Based2026.06 | 0 | 0 | 0 | |
| LC-Rec-SFT-Only-14BCategory=Itemic Token-Based2026.06 | 0 | 0 | 0 |