Textual Long-Horizon Complex Reasoning on GAIA Text
81.6Overall ScoreMiniMax-M2.7
Evaluation results
| Method | Links | |
|---|---|---|
| MiniMax-M2.7Size=230B, Train=-2026.06 | 81.6 | |
| REDSearcherSize=30B, Train=Mid/SFT/RL2026.06 | 80.1 | |
| Doubao-2.0-ProSize=-, Train=-2026.06 | 78.6 | |
| Qwen3.5-397BSize=397B, Train=-2026.06 | 73.8 | |
| S1-DeepResearchSize=32B, Train=SFT2026.06 | 72.8 | |
| Claude-4.6-Sonnet-ThinkingSize=-, Train=-2026.06 | 71.8 | |
| Gemini-3.1-Pro-PreviewSize=-, Train=-2026.06 | 70.9 | |
| Tongyi-DeepResearchSize=30B, Train=Mid/SFT/RL2026.06 | 70.9 | |
| Kimi-K2.5Size=1T, Train=-2026.06 | 68.9 | |
| GPT-5.2Size=-, Train=-2026.06 | 68 | |
| GLM-5Size=744B, Train=-2026.06 | 68 | |
| OpenResearcherSize=30B, Train=SFT2026.06 | 64.1 | |
| DeepSeek-V3.2Size=671B, Train=-2026.06 | 63.5 | |
| Qwen3-235BSize=235B, Train=-2026.06 | 33 | |
| Qwen3-32BSize=32B, Train=-2026.06 | 30.2 |