BENCHMARK EXPLORER · reasoning
Humanity’s Last Exam
Expert-written questions spanning many academic disciplines.
What it measures
Accuracy on the reported HLE evaluation.
Interpretation caveats
Text-only and multimodal evaluations are different. Question content is not reproduced here.
Original benchmarkPublished results
| # | Modalities | Compare | ||||
|---|---|---|---|---|---|---|
| 1 | moonshotai | InputTextImageOutputText | 262,144tokens | |||
| 2 | deepseek-ai | InputTextOutputText | 163,840tokens | |||
| 3 | miromind-ai | InputTextOutputNot documented | —Not documented | |||
| 4 | nvidia | InputTextOutputNot documented | —Not documented | |||
| 5 | miromind-ai | InputTextOutputNot documented | —Not documented | |||
| 6 | zai-org | InputTextOutputText | 204,800tokens | |||
| 7 | Qwen | InputTextImageVideoOutputText | 262,144tokens | |||
| 8 | Qwen | InputTextImageVideoOutputText | 262,144tokens | |||
| 9 | meituan-longcat | InputTextOutputNot documented | —Not documented | |||
| 10 | zai-org | InputTextOutputText | 204,800tokens | |||
| 11 | Qwen | InputTextImageVideoOutputText | 262,144tokens | |||
| 12 | moonshotai | InputTextOutputText | 262,144tokens | |||
| 13 | stepfun-ai | InputTextOutputText | 262,144tokens | |||
| 14 | Qwen | InputTextImageVideoOutputText | 262,144tokens | |||
| 15 | MiniMaxAI | InputTextOutputText | 204,800tokens | |||
| 16 | XiaomiMiMo | InputTextOutputNot documented | —Not documented | |||
| 17 | MiniMaxAI | InputTextOutputText | 204,800tokens | |||
| 18 | openbmb | InputTextOutputNot documented | —Not documented | |||
| 19 | nvidia | InputTextOutputNot documented | —Not documented | |||
| 20 | nvidia | InputTextOutputText | 262,144tokens | |||
| 21 | zai-org | InputTextOutputText | 200,000tokens | |||
| 22 | LGAI-EXAONE | InputTextOutputNot documented | —Not documented | |||
| 23 | Nanbeige | InputTextOutputNot documented | —Not documented | |||
| 24 | MiniMaxAI | InputTextOutputText | 204,800tokens | |||
| 25 | tiiuae | InputTextOutputNot documented | —Not documented | |||
| 26 | nvidia | InputTextOutputNot documented | —Not documented | |||
| 27 | HelpingAI | InputTextOutputNot documented | —Not documented | |||
| 28 | openai | InputTextOutputText | 131,072tokens | |||
| 29 | openai | InputTextOutputText | 131,072tokens | |||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | Anthropic | — | InputTextImageFileOutputText | 1,000,000tokens | ||
| — | nvidia | — | InputTextOutputImage | —Not documented | ||
| — | nvidia | — | InputTextOutputImage | —Not documented | ||
| — | deepseek-ai | — | InputTextOutputText | 64,000tokens | ||
| — | deepseek-ai | — | InputTextOutputText | 163,840tokens | ||
| — | deepseek-ai | — | InputTextOutputText | 163,840tokens | ||
| — | deepseek-ai | — | InputTextOutputText | 163,840tokens | ||
| — | — | InputTextOutputText | —Not documented | |||
| — | — | InputAudioFileImageTextVideoOutputText | 1,048,576tokens | |||
| — | — | InputTextImageVideoFileAudioOutputText | 1,048,576tokens | |||
| — | — | InputTextImageVideoFileAudioOutputText | 1,048,576tokens | |||
| — | — | InputAudioFileImageTextVideoOutputText | 1,048,576tokens | |||
| — | thinky | — | InputTextImageAudioOutputText | 524,288tokens |