Benchmark

MIABench

MIABench evaluates multimodal instruction alignment and following capabilities.

Modality
multimodal
Categories
multimodal, instruction_following, vision
Openness
unknown
Source
llm_stats
Reported scores
1

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Qwen3 VL 235B A22B ThinkingQwen0.9272025-09-22evidence

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.