LetTokenBurnBenchHub

Model

Qwen2.5-72B-Instruct

Alibaba · open weights

qwenlm.github.io

Knowledge & Reasoning

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
Balrog	16.2	average_progress	N/A	Independent	medium	balrogai.com
BBH	79.8	average	N/A	Independent	medium	arxiv.org
Colombian Legal Reliability Benchmark	67.2	accuracy	N/A	Independent	medium	arxiv.org
GPQA	49.0	accuracy	N/A	Independent	medium	arxiv.org
GPQA Diamond	49.1	accuracy	N/A	Independent	high	epoch.ai
HellaSwag	84.8	overall_accuracy	N/A	Independent	medium	arxiv.org
MMLU	85.3	exact_match	N/A	Independent	medium	arxiv.org
PHRBench	75.7	auroc	N/A	Independent	medium	arxiv.org
PIQA	82.6	score	N/A	Independent	medium	arxiv.org
Trivia QA	71.9	exact_match	N/A	Independent	medium	arxiv.org
WinoGrande	82.3	accuracy	N/A	Independent	medium	arxiv.org

Math

Benchmark	Score	Metric	Harness	Source	Confidence	Link
AIME24	23.3	accuracy	N/A	Independent	medium	arxiv.org
GraphDecide	100.0	accuracy	N/A	Independent	medium	arxiv.org

Coding

Benchmark	Score	Metric	Harness	Source	Confidence	Link
VHDL-REPOBENCH	53.1	bc_(pass@1)	N/A	Independent	medium	arxiv.org
Web-Bench	10.5	pass@1	N/A	Independent	medium	arxiv.org

Agentic / Tool Use

Agentic / Tool Use

Benchmark	Score	Metric	Harness	Source	Confidence	Link
TALK-Dem	86.3	success_rate	N/A	Independent	medium	arxiv.org
TheAgentCompany	5.7	resolved_rate	N/A	Independent	medium	github.com

Multimodal

Benchmark	Score	Metric	Harness	Source	Confidence	Link
MiniVer-V	0.6	f1	N/A	Independent	medium	arxiv.org
VisualNoiseQA	42.8	accuracy	bp	Independent	medium	arxiv.org

Reproduced by the community