Humanity's Last Exam (HLE)
A broad expert-level benchmark intended to test difficult questions across many domains and formats.
Multimodal Ranking
Signals for models that need to understand images, charts, video, and text together.
Main rankings require at least 3 of 7 included sourced benchmark rows.
Within the main ranking, rows are sorted by average normalized score; coverage is shown separately and used only as a tie-breaker.
Missing benchmark data is not treated as zero; models below the threshold appear under limited evidence.
| Rank | Model | Provider | Score | Coverage | Source | Source Type | Verified | Last Updated |
|---|
Limited Evidence
| Model | Provider | Score | Coverage | Freshest Source | Last Updated |
|---|---|---|---|---|---|
| OpenAI GPT-5.2 Thinking | OpenAI | 79.5 | 1/7 | Introducing GPT-5.2 Accessed 2026-06-11 | 2026-06-11 |
| Gemini 3.1 Pro | 63.9 | 2/7 | Scale Labs HLE Text Only leaderboard Accessed 2026-06-12 | 2026-06-12 | |
| Gemini 3 Pro | 59.4 | 2/7 | Scale Labs HLE Text Only leaderboard Accessed 2026-06-12 | 2026-06-12 | |
| OpenAI GPT-5.4 | OpenAI | 58.8 | 2/7 | Scale Labs HLE Text Only leaderboard Accessed 2026-06-12 | 2026-06-12 |
| Claude Opus 4.6 | Anthropic | 36.2 | 1/7 | Scale Labs HLE Text Only leaderboard Accessed 2026-06-12 | 2026-06-12 |
| DeepSeek-R1 | DeepSeek | 8.5 | 1/7 | Scale Labs HLE Text Only leaderboard Accessed 2026-06-12 | 2026-06-12 |
OSWorld / OSWorld-Verified
A computer-use benchmark for operating-system and GUI agents.
VisualWebArena
A visual web-agent benchmark extending WebArena with image-grounded web tasks.
MMMU-Pro
A harder multimodal academic benchmark derived from MMMU-style tasks.
Video-MME
A video understanding benchmark for multimodal models.
MathVista
A visual math reasoning benchmark for multimodal models.
CharXiv
A chart and figure understanding benchmark for scientific visual reasoning.