# HLE-Verified

> A verified and revised Humanity's Last Exam set that removes uncertain items and repairs fixable questions before evaluating frontier expert reasoning.

Canonical page: https://benchlm.ai/benchmarks/hleverified

- Category: [Knowledge](/knowledge)
- Last updated: October 17, 2026

## About HLE-Verified

- Year: 2026
- Tasks: 1,811 verified or revised expert questions
- Format: Full-set accuracy
- Difficulty: Frontier multidisciplinary expert reasoning
- Paper: [HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam](https://arxiv.org/abs/2602.13964)

HLE-Verified contains 1,811 certified items and separates 689 uncertain original HLE items. Google's Gemini 3.8 evaluation reports full-set accuracy as display-only scientific-reasoning evidence; it does not overwrite the existing HLE lanes.

HLE-Verified is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (6 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | Google | 54.9% |
| 2 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | OpenAI | 54.5% |
| 3 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | 54.4% |
| 4 | [Gemini 3.7 Flash](/models/gemini-3-7-flash) | Google | 53.6% |
| 5 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | OpenAI | 51.1% |
| 6 | [Claude Sonnet 5](/models/claude-sonnet-5) | Anthropic | 31.0% |

## FAQ

### What does HLE-Verified measure?

A verified and revised Humanity's Last Exam set that removes uncertain items and repairs fixable questions before evaluating frontier expert reasoning.

### Which model scores highest on HLE-Verified?

Gemini 3.8 Flash by Google currently leads with a score of 54.9% on HLE-Verified.

### How many models are evaluated on HLE-Verified?

6 AI models have published results on HLE-Verified in the BenchLM catalog.

### Does HLE-Verified affect BenchLM's overall score?

Not directly. HLE-Verified is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Compare Top Models on HLE-Verified

- [Gemini 3.8 Flash vs GPT-5.6 Sol](/compare/gemini-3-8-flash-vs-gpt-5-6-sol)
- [GPT-5.6 Sol vs Claude Opus 5](/compare/claude-opus-5-vs-gpt-5-6-sol)
- [Claude Opus 5 vs Gemini 3.7 Flash](/compare/claude-opus-5-vs-gemini-3-7-flash)
- [Gemini 3.7 Flash vs GPT-5.6 Terra](/compare/gemini-3-7-flash-vs-gpt-5-6-terra)