LetTokenBurnBenchHub

Model

INTELLECT-1-Instruct

Prime Intellect,Hugging Face,Arcee AI

Knowledge & Reasoning

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
BBH	34.9	average	N/A	Self-reported	medium	arxiv.org
HellaSwag	71.4	overall_accuracy	N/A	Self-reported	medium	arxiv.org
MMLU	49.9	exact_match	N/A	Self-reported	medium	arxiv.org
WinoGrande	65.8	accuracy	N/A	Self-reported	medium	arxiv.org

Math

Benchmark	Score	Metric	Harness	Source	Confidence	Link
GSM8K	38.6	exact_match	N/A	Self-reported	medium	arxiv.org

Reproduced by the community