Long-Context Reasoning on Oolong-Synth
78.41AccuracyGemini-3.0-pro
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini-3.0-pro2026.03 | 78.41 | |
| Deepseek-v3.12026.03 | 51.9 | |
| Deepseek-R1-Distill-Qwen-32B + TableLong2026.03 | 51.41 | |
| Deepseek-R1-Distill-Qwen-14B + TableLong2026.03 | 45.95 | |
| Qwen3-32B + TableLong2026.03 | 42.31 | |
| Deepseek-R1-Distill-Qwen-14B2026.03 | 42.05 | |
| Qwen2.5-32B-Instruct2026.03 | 41.03 | |
| Deepseek-R1-Distill-Qwen-32B2026.03 | 39.43 | |
| Qwen2.5-32B-Instruct + TableLong2026.03 | 35.5 | |
| Qwen3-32B2026.03 | 31.66 | |
| Qwen-Long-L12026.03 | 27.37 |