LetTokenBurnBenchHub

Model

Qwen2.5-7B-Instruct

Alibaba · open weights

qwenlm.github.io

Knowledge & Reasoning

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
Balrog	7.8	average_progress	N/A	Independent	medium	balrogai.com
GPQA Diamond	35.5	accuracy	N/A	Independent	high	epoch.ai
LOGIC	0.8	f1	N/A	Independent	medium	arxiv.org
MMLU	72.9	exact_match	N/A	Independent	medium	crfm.stanford.edu
PHRBench	64.5	auroc	N/A	Independent	medium	arxiv.org
ViLegalExpert	71.3	score	N/A	Independent	medium	arxiv.org

Math

Benchmark	Score	Metric	Harness	Source	Confidence	Link
DICE-Bench	86.5	success_rate	dice-cost	Independent	medium	arxiv.org

Coding

Benchmark	Score	Metric	Harness	Source	Confidence	Link
AlgoREval	37.4	accuracy	N/A	Independent	medium	arxiv.org

Agentic / Tool Use

Agentic / Tool Use

Benchmark	Score	Metric	Harness	Source	Confidence	Link
MADS-Bench	39.8	success_rate	N/A	Independent	medium	arxiv.org
NAQD-Env	0.5	withdrawal_recall	naqd-env	Independent	medium	arxiv.org
PFM	0.2	clean-retrieval_rate	N/A	Independent	medium	arxiv.org
TALK-Dem	28.7	success_rate	N/A	Independent	medium	arxiv.org

Multimodal

Benchmark	Score	Metric	Harness	Source	Confidence	Link
VisualNoiseQA	37.3	accuracy	bp	Independent	medium	arxiv.org

Long Context

Benchmark	Score	Metric	Harness	Source	Confidence	Link
CICM	12.3	accuracy	N/A	Independent	medium	arxiv.org

Instruction Following

Instruction Following

Benchmark	Score	Metric	Harness	Source	Confidence	Link
Intent-Eval	91.7	accuracy	N/A	Independent	medium	arxiv.org

Safety & Alignment

Safety & Alignment

Benchmark	Score	Metric	Harness	Source	Confidence	Link
LPHB	0.6	kappa	N/A	Independent	medium	arxiv.org

Reproduced by the community