BENCHMARK EXPLORER · math
FrontierMath Tier 4 v2
FrontierMath Tier 4 v2 results published in Epoch AI’s benchmark data.
What it measures
Mean score in the named Epoch AI evaluation; reasoning configurations remain separate.
Interpretation caveats
Scores apply to the reported benchmark and exact model configuration. Other evaluators, dataset revisions and tool settings may differ.
Original benchmarkPublished results
| # | Modalities | Compare | ||||
|---|---|---|---|---|---|---|
| 1 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 2 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 2 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 2 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 3 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 4 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 5 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 6 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 7 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 8 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 8 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 9 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 9 | OpenAI | InputTextOutputText | —Not documented | |||
| 10 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 10 | OpenAI | InputTextOutputText | —Not documented | |||
| 11 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 11 | Google DeepMind | InputAudioFileImageTextVideoOutputText | 1,048,576tokens | |||
| 12 | Google DeepMind | InputTextOutputText | —Not documented | |||
| 13 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 14 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 15 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 15 | Meta AI | InputTextImageVideoFileAudioOutputText | 1,048,576tokens | |||
| 16 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 17 | OpenAI | InputTextImageFileOutputText | 1,050,000tokens | |||
| 18 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 18 | OpenAI | InputFileImageTextOutputText | 1,050,000tokens | |||
| 19 | OpenAI | InputTextImageFileOutputText | 1,050,000tokens | |||
| 20 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 20 | DeepSeek | InputTextImageOutputText | 1,048,576tokens | |||
| 20 | Meta AI | InputTextImageVideoFileOutputText | 1,048,576tokens | |||
| 20 | Alibaba | InputTextOutputText | —Not documented | |||
| 21 | OpenAI | InputImageTextFileOutputText | 400,000tokens | |||
| 22 | Meta AI | InputTextImageVideoFileOutputText | 1,048,576tokens | |||
| 23 | Moonshot | InputTextImageVideoOutputText | 1,048,576tokens | |||
| 24 | Google DeepMind | InputTextImageVideoFileAudioOutputText | 1,048,576tokens | |||
| 25 | Alibaba | InputTextOutputText | 1,000,000tokens | |||
| 25 | Alibaba | InputTextImageVideoOutputText | 1,000,000tokens | |||
| 26 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 26 | xAI | InputTextImageFileOutputText | 500,000tokens | |||
| 26 | Meta AI | InputTextImageVideoFileAudioOutputText | 1,048,576tokens | |||
| 27 | OpenAI | InputFileImageTextOutputText | 400,000tokens | |||
| 28 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 28 | Z.ai (Zhipu AI) | InputTextOutputText | 1,048,576tokens | |||
| 28 | Z.ai (Zhipu AI) | InputTextOutputText | 1,048,576tokens | |||
| 29 | Anthropic | InputTextImageFileOutputText | 1,000,000tokens | |||
| 29 | DeepSeek | InputTextOutputText | 1,048,576tokens | |||
| 29 | DeepSeek | InputTextImageOutputText | 1,048,576tokens | |||
| 29 | Google DeepMind | InputAudioFileImageTextVideoOutputText | 1,048,576tokens | |||
| 29 | Google DeepMind | InputTextImageVideoFileAudioOutputText | 1,048,576tokens | |||
| 29 | Mistral AI | InputTextImageOutputText | 1,048,576tokens |