# GPQA Diamond Contamination Audit

> GPQA Diamond scores 4.4/10 (Tier C, High 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/gpqa-diamond/
- **Benchmark ID:** `gpqa-diamond`
- **Hub:** Artificial Analysis / NYU
- **Category:** Science
- **Domains:** Science
- **Skills:** Reasoning, Knowledge Work
- **BenchAudit Score:** 4.4/10 (Tier C - High Risk)
- **Last Audited:** 2026-11-17 (Ref 2026-11)

## Factor Scores

| Factor | Score (/10) | Weight |
| --- | ---: | ---: |
| Privacy | 2 | 40% |
| Originality | 10 | 25% |
| Freshness | 0 | 20% |
| Grading | 7.3 | 15% |

## Primary-Source Evidence

### Privacy
> "contains the data files, and is password-proteced with this password: deserted-untie-orchid"
- Source: https://github.com/idavidrein/gpqa

### Originality
> "198 questions covering biology, physics, and chemistry - we test the GPQA Diamond subset of the full GPQA dataset (448 questions total), which was defined by the original authors as the highest quality subset, where both experts answer correctly and the majority of non-experts answer incorrectly"
- Source: https://artificialanalysis.ai/methodology/intelligence-benchmarking

### Freshness
> "arXiv:2311.12022 (cs) [Submitted on 20 Nov 2023]"
- Source: https://arxiv.org/abs/2311.12022

### Grading
> "4 option multiple choice format Regex-based answer extraction with pass@1 scoring (prompt and regex below)"
- Source: https://artificialanalysis.ai/methodology/intelligence-benchmarking
