# Chartography Contamination Audit

> Chartography scores 4.1/10 (Tier C, High Risk) on BenchAudit as of November 2026, reflecting publicly downloadable plaintext test files and standardized independent third-party grading.

- **Canonical URL:** https://benchaudit.org/benchmarks/chartography/
- **Benchmark ID:** `chartography`
- **Hub:** Surge AI
- **Category:** Knowledge Work
- **Domains:** Science, Finance, Medical, Business
- **Skills:** Multimodal, Reasoning, Knowledge Work
- **BenchAudit Score:** 4.1/10 (Tier C - High Risk)
- **Last Audited:** 2026-11-19 (Ref 2026-11)

## Factor Scores

| Factor | Score (/10) | Weight |
| --- | ---: | ---: |
| Privacy | 0 | 40% |
| Originality | 7.2 | 25% |
| Freshness | 6 | 20% |
| Grading | 7.3 | 15% |

## Primary-Source Evidence

### Privacy
> "We release all tasks, images, provenance metadata, and evaluation code through the public dataset and code repositories, and will maintain the public leaderboard as new models launch."
- Source: https://arxiv.org/html/2608.10677v1

### Originality
> "Each task was authored by a practitioner in the relevant field, accompanied by an explicit walkthrough of the visual and computational steps leading to the answer, and independently verified by three additional domain experts."
- Source: https://arxiv.org/html/2608.10677v1

### Freshness
> "Table 3 reports mean pass@1 for all 30 frontier model configurations in the launch snapshot (July 2026)"
- Source: https://arxiv.org/html/2608.10677v1

### Grading
> "Each model receives a single user message containing the task prompt followed by the chart image inline: no system prompt, no few-shot examples, no image normalization, and no tools (no code execution, cropping, or zooming)."
- Source: https://arxiv.org/html/2608.10677v1
