Maths · %
AIME (OTIS Mock AIME 2024–2025)
AIME-style competition maths (integer answers 0–999). We use Epoch AI's runs on the OTIS Mock AIME 2024–2025 papers, which are less exposed than past official AIME papers.
Active Contamination risk: Medium 265 results · 200 models Updated Oct 22, 2026 Newest result: 2026-10-19 Official leaderboard ↗
What it measures
Accuracy, averaged over runs.
Limitations
Few questions, so results are noisy. Vendor 'AIME 2025' claims in launch posts are not comparable with these numbers.
Contamination risk: Medium
Mock papers are public after release.
Leaderboard
Top 100 of 200
| # | Model | Provider | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|---|---|
| 1 | Gemini 4 Argon | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Oct 13, 2026 | |
| 2 | GPT-6.1 Sol | OpenAI | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Sep 29, 2026 |
| 3 | Claude Sonnet 5.5 | Anthropic | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Sep 29, 2026 |
| 4 | GPT-6 Sol | OpenAI | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Sep 22, 2026 |
| 5 | Claude Opus 5.5 | Anthropic | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Sep 22, 2026 |
| 6 | Qwen3.8 Max (0902) | Alibaba (Qwen) | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Sep 2, 2026 |
| 7 | Claude Fable 5.1 | Anthropic | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Sep 1, 2026 |
| 8 | GPT-6 Astra | OpenAI | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Aug 30, 2026 |
| 9 | Claude Fable 5 | Anthropic | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Aug 6, 2026 |
| 10 | GPT-5.6 Sol | OpenAI | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Jul 9, 2026 |
| 11 | GPT-5.5 Pro | OpenAI | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Apr 24, 2026 |
| 12 | GPT-5.5 | OpenAI | 100.0% | 1 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Apr 24, 2026 |
| 13 | GPT-5.6 Terra | OpenAI | 99.7% | 13 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Jul 9, 2026 |
| 14 | DeepSeek V4.1 Flash 1017OW | DeepSeek | 99.4% | 15 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Oct 19, 2026 |
| 15 | Muse Wonder | Meta | 99.4% | 15 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Oct 12, 2026 |
| 16 | Qwen 3.8 Max | Alibaba (Qwen) | 99.4% | 15 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Aug 4, 2026 |
| 17 | Muse Spark 1.3 | Meta | 99.2% | 18 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Sep 17, 2026 |
| 18 | Grok 4.6 | xAI | 99.2% | 18 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Aug 14, 2026 |
| 19 | Claude Haiku 5.5 | Anthropic | 98.9% | 20 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Oct 7, 2026 |
| 20 | GPT-6 Luna | OpenAI | 98.9% | 20 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Sep 22, 2026 |
| 21 | Gemini 3.8 Flash | 98.9% | 20 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Sep 2, 2026 | |
| 22 | Claude Opus 5 | Anthropic | 98.9% | 20 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Jul 24, 2026 |
| 23 | DeepSeek V4 Pro 0813OW | DeepSeek | 98.6% | 24 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Aug 18, 2026 |
| 24 | Mistral Large 4 | Mistral AI | 98.3% | 25 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Oct 12, 2026 |
| 25 | DeepSeek V4.1 FlashOW | DeepSeek | 98.3% | 25 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Oct 8, 2026 |
| 26 | GPT-5.6 Luna | OpenAI | 98.3% | 25 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Jul 9, 2026 |
| 27 | Muse Spark OpenOW | Meta | 98.1% | 28 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Oct 15, 2026 |
| 28 | Grok 4.7 | xAI | 98.1% | 28 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Sep 22, 2026 |
| 29 | Claude Opus 4.7 | Anthropic | 97.8% | 30 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Apr 17, 2026 |
| 30 | GPT-5.4 | OpenAI | 97.8% | 31 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Jul 15, 2026 |
| 31 | Grok 4.5 | xAI | 97.8% | 31 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Jul 8, 2026 |
| 32 | Gemini 3.7 Flash | 97.2% | 36 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Aug 14, 2026 | |
| 33 | Kimi K3OW | Moonshot AI | 97.2% | 36 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Jul 16, 2026 |
| 34 | DeepSeek-V4-ProOW | DeepSeek | 96.7% | 39 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Jun 17, 2026 |
| 35 | Kimi K2.6OW | Moonshot AI | 96.1% | 40 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | May 2, 2026 |
| 36 | GPT-5.2 | OpenAI | 96.1% | 40 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Dec 11, 2025 |
| 37 | Qwen3.7-Max | Alibaba (Qwen) | 95.6% | 43 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 38 | Kimi K2.7 CodeOW | Moonshot AI | 95.6% | 43 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 39 | Gemini 3.1 Pro | 95.6% | 43 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 6, 2026 | |
| 40 | Gemini 3 Flash | 95.6% | 43 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 6, 2026 | |
| 41 | Claude Sonnet 5 | Anthropic | 94.7% | 51 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Jul 1, 2026 |
| 42 | DeepSeek V4 Flash 0731OW | DeepSeek | 94.4% | 52 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Aug 2, 2026 |
| 43 | Claude Opus 4.6 | Anthropic | 94.4% | 52 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: 64k thinking | Independent run | Feb 6, 2026 |
| 44 | GLM-5.3-FlashOW | Zhipu AI (Z.ai) | 93.9% | 54 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Aug 26, 2026 |
| 45 | GLM-5.1OW | Zhipu AI (Z.ai) | 93.3% | 56 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 10, 2026 |
| 46 | Qwen 3.6 Plus | Alibaba (Qwen) | 93.3% | 56 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: 2026-04-02 | Independent run | Aug 7, 2026 |
| 47 | Qwen3.7-Plus | Alibaba (Qwen) | 93.3% | 56 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 48 | Grok 4.3 Beta | xAI | 93.3% | 56 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Jun 17, 2026 |
| 49 | Grok 4.20 | xAI | 92.2% | 63 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Jul 13, 2026 |
| 50 | Kimi K2.5OW | Moonshot AI | 92.2% | 64 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: Fireworks | Independent run | Feb 2, 2026 |
| 51 | Gemini 3 Pro | 91.4% | 65 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Nov 19, 2025 | |
| 52 | GPT-5 | OpenAI | 91.4% | 65 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Oct 29, 2025 |
| 53 | GLM-5.3OW | Zhipu AI (Z.ai) | 91.1% | 67 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: max | Independent run | Aug 24, 2026 |
| 54 | Qwen 3.6 Max (Preview) | Alibaba (Qwen) | 91.1% | 67 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: Preview | Independent run | Aug 7, 2026 |
| 55 | Qwen3.6 27BOW | Alibaba (Qwen) | 91.1% | 67 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 56 | Inkling-SmallOW | Thinking Machines | 90.0% | 71 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Aug 14, 2026 |
| 57 | Muse Spark | Meta | 88.9% | 72 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Apr 8, 2026 |
| 58 | GPT-5.4 Mini | OpenAI | 88.9% | 73 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Aug 7, 2026 |
| 59 | Qwen3.5 397B-A17BOW | Alibaba (Qwen) | 88.9% | 73 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 60 | Gemini 3.5 Flash | 88.9% | 73 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 6, 2026 | |
| 61 | InklingOW | Thinking Machines | 88.9% | 73 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Aug 5, 2026 |
| 62 | gpt-oss-120bOW | OpenAI | 88.9% | 73 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Dec 11, 2025 |
| 63 | GPT-5.1 | OpenAI | 88.6% | 79 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Nov 13, 2025 |
| 64 | DeepSeek-V3.2OW | DeepSeek | 87.8% | 80 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: Thinking | Independent run | Dec 16, 2025 |
| 65 | GPT-5.4 Nano | OpenAI | 87.8% | 81 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Apr 14, 2026 |
| 66 | Nemotron 3 UltraOW | NVIDIA | 86.7% | 84 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 10, 2026 |
| 67 | Qwen 3.5 Plus (hosted 397B-A17B) | Alibaba (Qwen) | 86.7% | 84 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 68 | Qwen 3.6 35B-A3BOW | Alibaba (Qwen) | 86.7% | 84 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 69 | Qwen3.7 Flash | Alibaba (Qwen) | 86.7% | 84 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 70 | Qwen3-235B-A22B-Thinking (Jul 2025)OW | Alibaba (Qwen) | 86.7% | 84 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Dec 10, 2025 |
| 71 | GPT-5 mini | OpenAI | 86.7% | 84 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Oct 30, 2025 |
| 72 | Claude Opus 4.5 | Anthropic | 86.1% | 91 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: 32k thinking | Independent run | Nov 24, 2025 |
| 73 | Claude Sonnet 4.6 | Anthropic | 85.8% | 92 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: 32k thinking | Independent run | Feb 20, 2026 |
| 74 | Gemini 2.5 Pro (Jun 2025) | 84.7% | 94 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Nov 16, 2025 | |
| 75 | o3 | OpenAI | 84.4% | 95 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: medium | Independent run | Aug 7, 2026 |
| 76 | Qwen 3.6 Flash | Alibaba (Qwen) | 84.4% | 95 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: 2026-04-16 | Independent run | Aug 7, 2026 |
| 77 | Qwen 3.5 Flash (hosted 35B-A3B) | Alibaba (Qwen) | 84.4% | 95 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 |
| 78 | Claude Opus 4.8 | Anthropic | 84.4% | 95 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 6, 2026 |
| 79 | Grok 4 | xAI | 84.0% | 101 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Jul 9, 2025 |
| 80 | GLM-4.7OW | Zhipu AI (Z.ai) | 83.3% | 103 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Jan 29, 2026 |
| 81 | Kimi K2 ThinkingOW | Moonshot AI | 83.1% | 104 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Nov 13, 2025 |
| 82 | Gemma 4 26B A4BOW | 82.2% | 105 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 28, 2026 | |
| 83 | o4-mini | OpenAI | 81.7% | 107 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Apr 16, 2025 |
| 84 | GPT-5 nano | OpenAI | 81.1% | 109 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Oct 31, 2025 |
| 85 | Gemini 3.6 Flash | 80.0% | 110 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 7, 2026 | |
| 86 | GLM-5OW | Zhipu AI (Z.ai) | 80.0% | 110 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Feb 12, 2026 |
| 87 | Claude Sonnet 4.5 | Anthropic | 77.8% | 115 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: 59k thinking | Independent run | Oct 28, 2025 |
| 88 | o3-mini | OpenAI | 76.9% | 117 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Feb 27, 2025 |
| 89 | Gemma 4 31B ITOW | 73.3% | 120 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 6, 2026 | |
| 90 | o1 | OpenAI | 73.3% | 120 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Jul 15, 2026 |
| 91 | Qwen3-Max | Alibaba (Qwen) | 73.3% | 120 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Oct 6, 2025 |
| 92 | Gemini 2.5 Flash (Apr 2025) | 73.1% | 125 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: Apr 2025 | Independent run | Apr 22, 2025 | |
| 93 | Gemini 2.5 Flash (May 2025) | 70.8% | 128 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | May 20, 2025 | |
| 94 | Qwen3-30B-A3B-Thinking (Jul 2025)OW | Alibaba (Qwen) | 70.3% | 129 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 30, 2026 |
| 95 | Qwen3.5-35B-A3BOW | Alibaba (Qwen) | 70.0% | 130 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 30, 2026 |
| 96 | Claude Opus 4.1 | Anthropic | 68.9% | 131 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 5, 2025 |
| 97 | Claude Sonnet 4 | Anthropic | 68.9% | 131 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | May 23, 2025 |
| 98 | GPT-5.5 Instant | OpenAI | 68.1% | 136 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 2, 2026 |
| 99 | seed-oss-36b-instruct | ByteDance | 67.5% | 137 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 28, 2026 |
| 100 | Claude Haiku 4.5 | Anthropic | 66.7% | 138 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Oct 22, 2025 |
Best score over time
Each step is a new best result in our data, by evaluation date.
- 2025-02-2563.9%o3-mini
- 2025-02-2776.9%o3-mini
- 2025-04-1683.9%o3
- 2025-07-0984.0%Grok 4
- 2025-08-0787.2%GPT-5
- 2025-10-2991.4%GPT-5
- 2025-12-1196.1%GPT-5.2
- 2026-04-1797.8%Claude Opus 4.7
- 2026-04-24100.0%GPT-5.5 Pro
Data sources & licences
| Source | Licence / terms | How we reuse it | Last fetched |
|---|---|---|---|
| Epoch AI Benchmarking Hub Epoch AI, 'Capabilities & benchmarking', https://epoch.ai/benchmarks (CC BY 4.0) | CC BY 4.0 | Open licence – full data with attribution | Oct 22, 2026 |
FAQ
What does AIME (OTIS Mock AIME 2024–2025) measure?
AIME-style competition maths (integer answers 0–999). We use Epoch AI's runs on the OTIS Mock AIME 2024–2025 papers, which are less exposed than past official AIME papers. Accuracy, averaged over runs.
How reliable is AIME (OTIS Mock AIME 2024–2025)?
Few questions, so results are noisy. Vendor 'AIME 2025' claims in launch posts are not comparable with these numbers. Contamination risk: Medium – Mock papers are public after release.