LetTokenBurnBenchHub

Model

Qwen2.5-VL-7B

Alibaba · open weights

qwenlm.github.io

Agentic / Tool Use

Agentic / Tool Use

Benchmark	Score	Metric	Harness	Source	Confidence	Link
PlaySuite	0.8	mean_progress_score_(no_history)	N/A	Independent	medium	arxiv.org

Multimodal

Benchmark	Score	Metric	Harness	Source	Confidence	Link
4MT-VLM	36.0	accuracy	N/A	Independent	medium	arxiv.org
BrainTRACE	33.2	accuracy	N/A	Independent	medium	arxiv.org
Counterfactual Physical Consistency Benchmark	95.3	bin._mf1	lora	Independent	medium	arxiv.org
CROSS-Bench	66.3	accuracy	N/A	Independent	medium	arxiv.org
DataVista	43.9	accuracy	N/A	Independent	medium	arxiv.org
EgoTools-Bench	40.1	egotools-bench_overall	N/A	Independent	medium	arxiv.org
FG-Sudoku	0.2	accuracy	N/A	Independent	medium	arxiv.org
G-RSVQA	65.5	accuracy	N/A	Independent	medium	arxiv.org
ImplicitQA	42.8	average_accuracy	N/A	Independent	medium	arxiv.org
InstanceBench	42.2	[email protected]	N/A	Independent	medium	arxiv.org
NegT2IBench	58.4	spearman_ρ	N/A	Independent	medium	arxiv.org
OmniVCBench	26.6	accuracy	N/A	Independent	medium	arxiv.org
PhysicsLENS	0.7	plausibility_score	N/A	Independent	medium	arxiv.org
SpaceConflict	24.4	accuracy	N/A	Independent	medium	arxiv.org
VisualNoiseQA	46.9	accuracy	bp	Independent	medium	arxiv.org
VLM2-Bench	35.9	gc-mat	N/A	Independent	medium	arxiv.org

Safety & Alignment

Safety & Alignment

Benchmark	Score	Metric	Harness	Source	Confidence	Link
OR-Bench	54.9	accuracy	N/A	Independent	medium	arxiv.org

Reproduced by the community