Model

Qwen3-VL-30B-A3B

Alibaba · open weights

qwenlm.github.io

Agentic / Tool Use

Agentic / Tool Use
BenchmarkScoreMetricHarnessSourceConfidenceLink
Kolab0.5f1N/AIndependentmediumarxiv.org
PlanBench20.0accuracyN/AIndependentmediumarxiv.org

Multimodal

Multimodal
BenchmarkScoreMetricHarnessSourceConfidenceLink
EgoGears25.8accuracyN/AIndependentmediumarxiv.org
Game2Real-Bench21.8accuracyN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…