Humanity's Last Exam
- OpenAI
- Anthropic
- Reported in a later post or card
Results
| Model | Score | Source | Human-verified |
|---|---|---|---|
| Claude Haiku 5.57 Oct 2026 · Anthropic | 45.9% | Introducing Claude Haiku 5.5 | Not yet |
| Claude Sonnet 5.528 Sep 2026 · Anthropic | 56.9% | Claude Sonnet 5.5 System Cardthe launch post reports HLE with tools only | Not yet |
| Claude Opus 5.522 Sep 2026 · Anthropic | 64.4% | Claude Opus 5.5 System Cardthe launch post reports HLE with tools only | Not yet |
| Claude Opus 524 Jul 2026 · Anthropic | 56.3% | Introducing Claude Opus 5Source image | Not yet |
| Claude Opus 524 Jul 2026 · Anthropic | 56.6% | Claude Opus 5.5 System Cardreported later; another report, not charted | Not yet |
| Claude Sonnet 530 Jun 2026 · Anthropic | 43.2% | Introducing Claude Sonnet 5Source image | Not yet |
| Claude Sonnet 530 Jun 2026 · Anthropic | 43.1% | Claude Sonnet 5.5 System Cardreported later; another report, not charted | Not yet |
| Claude Opus 4.828 May 2026 · Anthropic | 49.8% | Introducing Claude Opus 4.8Source image | Not yet |
| Claude Opus 4.716 Apr 2026 · Anthropic | 46.9% | Introducing Claude Opus 4.7Source image | Not yet |
| GPT-5.45 Mar 2026 · OpenAI | 39.8% | Introducing GPT-5.4 | Not yet |
| Gemini 3.1 Pro19 Feb 2026 · Google | 44.4% | Gemini 3.1 Pro: A smarter model for your most complex tasksThinking (High)Source image | Not yet |
| Claude Opus 4.65 Feb 2026 · Anthropic | 40.0% | Claude Opus 4.6Source image | Not yet |
| GPT-5.211 Dec 2025 · OpenAI | 34.5% | Introducing GPT-5.2 | Not yet |
| Claude Opus 4.524 Nov 2025 · Anthropic | 30.8% | Claude Opus 4.5 System Cardreasoning only, without tools; Claude Sonnet 4.5 as grader (section 2.16) | Not yet |
| Gemini 3 Pro18 Nov 2025 · Google | 37.5% | Start building with Gemini 3Source image | Not yet |
| Claude Haiku 4.515 Oct 2025 · Anthropic | 10.2% | Introducing Claude Haiku 5.5reported later | Not yet |
| Claude Sonnet 4.529 Sep 2025 · Anthropic | 17.7% | Claude Opus 4.5 System Cardreported later; Anthropic notes decontamination improved with Opus 4.5, which may affect this score | Not yet |
Benchmark
lastexam.aiHigher is better