Math on IMO-ANSWERBENCH (Score)
89.3ScoreSeed2.0 Pro
Evaluation results
| Method | Links | |
|---|---|---|
| Seed2.0 Protool=no tool2026.06 | 89.3 | |
| GPT-5.2 Hightool=no tool2026.06 | 86.6 | |
| Gemini-3-pro Hightool=no tool2026.06 | 83.3 | |
| Claude-Opus-4.5tool=no tool2026.06 | 72.6 | |
| Claude-Sonnet-4.5tool=no tool2026.06 | 60.7 | |
| ZAYA1-8BActive / Total Parameters=0.7B / 8.0B, Sampling Settings=T=1.0, top-p=0.95, top-k disabled2026.05 | 59.3 | |
| Qwen3-235B-A22B Instruct-2507Architecture=MoE, Total Params=235B, Activated Params=22B2026.01 | 53.8 | |
| Qwen3-4B-Thinking-2507Active / Total Parameters=4.0B / 4.0B, Sampling Settings=Recommended sampling settings2026.05 | 51.6 | |
| Qwen3.5-4BActive / Total Parameters=4.0B / 4.0B, Sampling Settings=Recommended sampling settings2026.05 | 48.7 | |
| DeepSeek-V3.2Mode=Non-Reasoning, Architecture=MoE, Total Params=671B, Activated Params=37B2026.01 | 45.9 | |
| K-EXAONEMode=Non-Reasoning, Architecture=MoE, Total Params=236B, Activated Params=23B2026.01 | 40 | |
| EXAONE 4.0Mode=Non-Reasoning, Architecture=Dense, Total Params=32B, Activated Params=32B2026.01 | 35.8 | |
| Gemma-4-E4B-itActive / Total Parameters=4.0B / 8.0B, Sampling Settings=Recommended sampling settings2026.05 | 27.3 | |
| STEP3-VL-10BNumber of Parameters=10B2026.01 | 0.6212 | |
| MiMo-VL RL-2508Number of Parameters=7B2026.01 | 0.4844 | |
| Qwen3-VL ThinkingNumber of Parameters=8B2026.01 | 0.3869 | |
| InternVL 3.5Number of Parameters=8B2026.01 | 0.35 | |
| GLM-4.6V FlashNumber of Parameters=9B2026.01 | 0.2262 |