# HLE-Diamond Contamination Audit

> HLE-Diamond scores 5.8/10 (Tier B, Moderate Risk) on BenchAudit as of November 2026, reflecting auto-approved click-through or encrypted public access and standardized independent third-party grading.

- **Canonical URL:** https://benchaudit.org/benchmarks/hle-diamond/
- **Benchmark ID:** `hle-diamond`
- **Hub:** Artificial Analysis / CAIS / Scale AI
- **Category:** General
- **Domains:** Science, Math, Humanities, Medical
- **Skills:** Reasoning, Knowledge Work, Multimodal
- **BenchAudit Score:** 5.8/10 (Tier B - Moderate Risk)
- **Last Audited:** 2026-11-17 (Ref 2026-11)

## Factor Scores

| Factor | Score (/10) | Weight |
| --- | ---: | ---: |
| Privacy | 2 | 40% |
| Originality | 10 | 25% |
| Freshness | 7 | 20% |
| Grading | 7.3 | 15% |

## Primary-Source Evidence

### Privacy
> "This repository is publicly accessible, but you have to accept the conditions to access its files and content"
- Source: https://huggingface.co/datasets/cais/hle-diamond

### Originality
> "HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval."
- Source: https://arxiv.org/abs/2501.14249

### Freshness
> "Center for AI Safety and Scale AI September 22, 2026"
- Source: https://lastexam.ai/blog/hle-diamond

### Grading
> "We grade each answer with the official HLE judge prompt and response format, using GPT-5.6 Luna (medium) as the judge, with pass@1 scoring."
- Source: https://artificialanalysis.ai/methodology/intelligence-benchmarking

## Top Frontier Models on HLE-Diamond

| Rank | Model | Creator | Percentile |
| ---: | --- | --- | ---: |
| #1 | [Claude Fable 5.2](https://benchaudit.org/models/claude-fable-5-2__max/) | Anthropic | 100.0p |
| #2 | [GPT-6.2 Astra](https://benchaudit.org/models/gpt-6-2-astra__max/) | OpenAI | 97.9p |
| #3 | [Gemini 4 Neon](https://benchaudit.org/models/gemini-4-neon__xhigh/) | Google | 95.8p |
| #4 | [Gemini 4 Carbon](https://benchaudit.org/models/gemini-4-carbon__high/) | Google | 93.8p |
| #5 | [Claude Opus 5.5](https://benchaudit.org/models/claude-opus-5-5__max/) | Anthropic | 91.7p |
| #6 | [GPT-6.2 Sol](https://benchaudit.org/models/gpt-6-2-sol__max/) | OpenAI | 89.6p |
| #7 | [GPT-6 Astra](https://benchaudit.org/models/gpt-6-astra__max/) | OpenAI | 87.5p |
| #8 | [Claude Fable 5.1](https://benchaudit.org/models/claude-fable-5-1__max/) | Anthropic | 85.4p |
| #9 | [Gemini 4 Hydrogen](https://benchaudit.org/models/gemini-4-hydrogen__xhigh/) | Google | 82.3p |
| #9 | [GPT-6.1 Sol](https://benchaudit.org/models/gpt-6-1-sol__max/) | OpenAI | 82.3p |
