Four-Factor Integrity Breakdown & Primary-Source Evidence
Access Model: Public Plaintext
"We use the same 161 tasks as Handshake AI's upstream VIALS implementation, covering 174 source images and 8 artifact domains. Each model makes 3 independent repeats per task."
Originality
Weight 25%
10/10
Reuse Profile: All Original (0% reuse)
"We introduce VIALS, a visual question-answering benchmark with 161 such interpretation tasks, spanning the types of artifacts examined throughout experimental workflows in the biotech industry (rather than polished figures from publications and textbooks)."
Freshness
Weight 20%
7/10
Initial Release: 2026-08Latest Refresh: 2026-10Rolling Updates: No
"Models must be rerun on these tasks. Two prompts were updated to reduce remaining ambiguity, one reference answer and one image were updated. Released 2026-10-30 as the `v2.0.0` tag."
Integrity: Independent Offline (Leak-Verified)Sandbox: offline isolated
"Models receive the question and image file paths in the prompt, with the original image files available in a sandbox for code execution with Python and no network access."
Frontier Model Standings on VIALS-AA
Top 15 active models tracked
Benchmark maintainers can embed this live SVG badge in GitHub READMEs or documentation:
[](https://benchaudit.org/benchmarks/vials-aa/)