IMO-AnswerBench
Updated Apr 24, 2026·1,144 views
- Problem Count
- 400
- Institution
- DeepMind
- Category
- Math and Reasoning
- Metrics
- Accuracy
- Language
- English
- Difficulty
- Mixed
Overview
IMO-AnswerBench is an AI benchmark used to evaluate model capabilities. Review its overview, metrics, official resources, and model leaderboard results on DataLearnerAI.
Related resources
Latest IMO-AnswerBench model rankings and full benchmark leaderboard
Browse the latest scores, model modes, release dates, and parameter sizes for IMO-AnswerBench.
Source: DataLearnerAI
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Model Mode Legend
License:
Origin:
Model release cutoff:
IMO-AnswerBench Rank
| Rank | Model | License | |||
|---|---|---|---|---|---|
 DeepSeek-V4-Pro Thinking Level · High | 89.80 | 2026-04-24 | 1600B | Free Commercial | |
DeepSeek-V4-Flash Thinking Level · High | 88.40 | 2026-04-24 | 284B | Free Commercial | |
DeepSeek-V4-Pro Thinking Level · High | 88.00 | 2026-04-24 | 1600B | Free Commercial | |
4 |  Step 3.5 Flash Thinking EnabledTools | 86.70 | 2026-02-02 | 196B | Free Commercial |
5 |  Kimi K2.6 Thinking Enabled | 86.00 | 2026-04-20 | 1000B | Free Commercial |
6 | Step 3.5 Flash Thinking Enabled | 85.40 | 2026-02-02 | 196B | Free Commercial |
7 | DeepSeek-V4-Flash Thinking Level · High | 85.10 | 2026-04-24 | 284B | Free Commercial |
8 |  Qwen3-Max-Thinking Thinking Enabled | 83.90 | 2026-01-26 | 1000B | Closed |
9 |  GLM 5.1 Thinking Enabled | 83.80 | 2026-03-27 | 75.4B | Free Commercial |
10 | Qwen 3.6 Plus Preview Thinking Enabled | 83.80 | 2026-03-31 | Unknown | Closed |
11 | GLM-5 Thinking Enabled | 82.50 | 2026-02-11 | 744B | Free Commercial |
12 | Kimi K2.5 Thinking Enabled | 81.80 | 2026-01-27 | 1000B | Free Commercial |
13 | Qwen3.5-397B-A17B Thinking Enabled | 80.90 | 2026-02-16 | 39.7B | Free Commercial |
14 | Qwen3.6-27B Thinking Enabled | 80.80 | 2026-04-22 | 27B | Free Commercial |
15 | Qwen3.6-35B-A3B Thinking Enabled | 78.90 | 2026-04-16 | 35B | Free Commercial |
16 | DeepSeek-V4-Flash Standard Mode | 41.90 | 2026-04-24 | 284B | Free Commercial |
17 | DeepSeek-V4-Pro Standard Mode | 35.30 | 2026-04-24 | 1600B | Free Commercial |