LetTokenBurnBenchHub

Model

Qwen3-VL-2B

Alibaba · open weights

qwenlm.github.io

Coding

Benchmark	Score	Metric	Harness	Source	Confidence	Link
VISTA	2.6	score	N/A	Independent	medium	arxiv.org

Multimodal

Benchmark	Score	Metric	Harness	Source	Confidence	Link
AgroGround	51.8	accuracy	zero-shot	Independent	medium	arxiv.org
FloodVL-Pix	0.4	clair	N/A	Independent	medium	arxiv.org
FoveDoc-Bench	0.2	strict_accuracy	N/A	Independent	medium	arxiv.org
OmniVCBench	21.7	accuracy	N/A	Independent	medium	arxiv.org
SpaceConflict	25.7	accuracy	N/A	Independent	medium	arxiv.org
ViSiP-Bench	59.4	par	N/A	Independent	medium	arxiv.org

Safety & Alignment

Safety & Alignment

Benchmark	Score	Metric	Harness	Source	Confidence	Link
JUPE	55.6	rejection_rate_drop	N/A	Independent	medium	arxiv.org

Reproduced by the community