CraftBench
Explore score history ↗CraftBench evaluates scientific figure generation across three figure types and four input conditions, with human-drawn targets and a referenced VLM judge for scoring.
- Released
- 2026-05-28
- Evaluates
- Agents, General AI, Vision & 3D
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
CraftBench paper and code
No reported scores are on record for this benchmark yet.