IMOAnswerBench
IMOAnswerBench - International Mathematical Olympiad answer evaluation benchmark.
Category
STEM & Reasoning
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for IMOAnswerBench
| Model | Score | Date | Verified |
|---|---|---|---|
| Qwen3.6-35B-A3B |
79.97%
|
16.04.2026 | Verified |
| Gemma4-31B-it |
73.72%
|
16.04.2026 | Unverified |
| Qwen3.5-35B-A3B |
76.99%
|
16.04.2026 | Unverified |
| Gemma4-26B-A4B |
73.44%
|
16.04.2026 | Unverified |
| Qwen3.6-27B |
82.67%
|
22.04.2026 | Verified |
| Qwen3.5-397B-A17B |
82.81%
|
22.04.2026 | Unverified |
| Claude 4.5 Opus |
87.22%
|
22.04.2026 | Unverified |
| Qwen3.5-27B |
81.39%
|
16.04.2026 | Unverified |
| NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 |
93.75%
|
04.06.2026 | Verified |
| MiniMax-M2.7 |
64.91%
|
04.06.2026 | Unverified |
| GLM-5.1 |
91.19%
|
04.06.2026 | Unverified |
| Kimi K2.6 |
97.30%
|
04.06.2026 | Unverified |
| DeepSeek-V4-Pro |
100.00%
|
04.06.2026 | Unverified |
| DeepSeek-V4-Flash |
93.47%
|
26.06.2026 | Verified |
| GLM-5.2 |
97.16%
|
17.06.2026 | Verified |
| Gemini 3.1 Pro |
82.95%
|
17.06.2026 | Unverified |
| Claude Opus 4.8 |
86.51%
|
17.06.2026 | Unverified |
| Qwen3.7-Max |
95.74%
|
17.06.2026 | Unverified |
| GPT-5.4 |
97.73%
|
23.08.2026 | Verified |
| Claude Opus 4.6 |
74.86%
|
23.08.2026 | Verified |
| Kimi K2.5 |
84.09%
|
23.08.2026 | Verified |
| Opus-4.6 |
74.86%
|
26.06.2026 | Verified |
| DeepSeek-V4-Pro |
95.45%
|
26.06.2026 | Verified |
| Spark-X2.5-4B |
73.30%
|
— | Verified |
| Spark-X2.5-1.7B |
32.39%
|
— | Verified |
| Qwen3.5-9B |
67.05%
|
— | Verified |
| Qwen3.5-4B |
65.20%
|
— | Verified |
| Gemma4-12B |
49.15%
|
— | Verified |
| Gemma4-E4B |
6.11%
|
— | Verified |
| Gemma4-E2B |
22.60
|
— | Verified |