IMOAnswerBench
We show this table for reference; we do not rank on it.
A challenging mathematical reasoning benchmark reported in DeepSeek-V4 model evaluations.
Benchmark score on IMOAnswerBench — October 29, 2026
We compile the IMOAnswerBench rows from provider self-reports. dots3-note Preview leads the table at 90.9%, followed by Qwen3.7 Max (90.0%) and DeepSeek V4 Pro 0813 (89.8%). We do not use these results to rank models overall.
dots3-note Preview
Dots Studio
Qwen3.7 Max
Alibaba
DeepSeek V4 Pro 0813
DeepSeek
14 modelsMathematicsCurrentDisplay onlyUpdated October 29, 2026
Benchmark score table (14 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | dots3-note PreviewDots Studio | 90.9% | Not reported | Open |
| 2 | Qwen3.7 MaxAlibaba | 90.0% | Not reported | Closed |
| 3 | DeepSeek V4 Pro 0813DeepSeek | 89.8% | Not reported | Open |
| 4 | DeepSeek V4 Flash 0731DeepSeek | 88.4% | Not reported | Open |
| 5 | DeepSeek V4 Pro (High)DeepSeek | 88.0% | Not reported | Open |
| 6 | Qwen3.7 PlusAlibaba | 86.0% | Not reported | Closed |
| 7 | DeepSeek V4 Flash (High)DeepSeek | 85.1% | Not reported | Open |
| 8 | Ling 3.0 FlashInclusionAI | 83.7% | Not reported | Open |
| 9 | K-EXAONE 2.0LG AI Research | 78.6% | Not reported | Open |
| 10 | Spark-X2.6-4BSparkLLM | 75.2% | Not reported | Open |
| 11 | ZAYA1-8BZyphra | 59.3% | Not reported | Open |
| 12 | LongCat-Flash-Lite-SparseMeituan | 49.4% | Not reported | Open |
| 13 | DeepSeek V4 FlashDeepSeek | 41.9% | Not reported | Open |
| 14 | DeepSeek V4 ProDeepSeek | 35.3% | Not reported | Open |
Among the reported IMOAnswerBench rows, dots3-note Preview is first at 90.9%. The third row is 1.1 points behind. The broader top-10 range is 15.7 points, so the table still separates the published systems.
14 models have been evaluated on IMOAnswerBench. The benchmark falls in the Mathematics category. IMOAnswerBench is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About IMOAnswerBench
Year
2026
Tasks
Advanced mathematical answer generation
Format
Pass@1 math benchmark
Difficulty
Olympiad-level mathematics
BenchLM stores IMOAnswerBench as a display-only provider-table row when exact values are published for frontier math comparisons.
Freshness and provenance
Version
IMOAnswerBench 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does IMOAnswerBench measure?
A challenging mathematical reasoning benchmark reported in DeepSeek-V4 model evaluations.
Which model scores highest on IMOAnswerBench?
dots3-note Preview by Dots Studio currently leads with a score of 90.9% on IMOAnswerBench.
How many models are evaluated on IMOAnswerBench?
14 AI models have published results on IMOAnswerBench in the BenchLM catalog.
Compare top models on IMOAnswerBench
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.