LetTokenBurnBenchHub

Model

Qwen3-VL-32B

Alibaba · open weights

qwenlm.github.io

Knowledge & Reasoning

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
GameCommBench	0.5	analytical_completeness	N/A	Independent	medium	arxiv.org

Agentic / Tool Use

Agentic / Tool Use

Benchmark	Score	Metric	Harness	Source	Confidence	Link
Kolab	0.3	f1	N/A	Independent	medium	arxiv.org

Multimodal

Benchmark	Score	Metric	Harness	Source	Confidence	Link
DataVista	57.3	accuracy	N/A	Independent	medium	arxiv.org
EgoExo-Next	37.0	accuracy	N/A	Independent	medium	arxiv.org
EMBER-Bench	26.5	accuracy	N/A	Independent	medium	arxiv.org
FloodVL-Pix	0.5	clair	N/A	Independent	medium	arxiv.org
HAPACT	21.4	[email protected]	N/A	Independent	medium	arxiv.org
Knossos	54.7	f1	N/A	Independent	medium	arxiv.org
NegT2IBench	76.5	spearman_ρ	N/A	Independent	medium	arxiv.org
PhysicsLENS	0.8	plausibility_score	N/A	Independent	medium	arxiv.org
R-GroundBench	71.9	accuracy	N/A	Independent	medium	arxiv.org
ScenePercept-Bench	0.5	score	N/A	Independent	medium	arxiv.org
SQUARE-Bench	90.0	overall	N/A	Independent	medium	arxiv.org
Visual Sanad	12.2	discrimination	N/A	Independent	medium	arxiv.org
WOVEN	71.4	accuracy	N/A	Independent	medium	arxiv.org

Safety & Alignment

Safety & Alignment

Benchmark	Score	Metric	Harness	Source	Confidence	Link
JUPE	22.2	rejection_rate_drop	N/A	Independent	medium	arxiv.org

Reproduced by the community