Benchmark

Blueprint-Bench 2

Blueprint-Bench 2 is an agentic spatial reasoning benchmark that evaluates a model's ability to understand, plan, and reason over architectural blueprints…

Modality
multimodal
Categories
multimodal, reasoning, agents
Openness
unknown
Source
llm_stats
Reported scores
2

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Claude Fable 5Anthropic0.3862026-06-09evidence
Gemini 3.5 FlashGoogle0.3362026-05-19evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.