Benchmark

ClawEval-MM

ClawEval-MM is the multimodal variant of ClawEval, evaluating agentic problem solving with visual inputs.

Modality
multimodal
Categories
multimodal, agents, vision
Openness
unknown
Source
llm_stats
Reported scores
4

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Qwen3.7-PlusQwen0.5572026-05-31evidence
Qwen3.8-27BQwen0.5692026-08-14evidence
Seed 2.1 ProByteDance0.512026-06-24evidence
Seed 2.1 TurboByteDance0.462026-06-24evidence

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.