HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/allenai/allenai/Olmo-3-32B-Think-SFT

allenai/Olmo-3-32B-Think-SFT

family olmo · 1 variants · unknown

provider
allenai
released
—
huggingface
https://huggingface.co/allenai/Olmo-3-32B-Think-SFT
best scores
MATH: 95.6AIME 2024: 73.5AIME 2025: 66.2BBH: 88.8AGIEval: 85.9HumanEval: 90.0MBPP: 66.7LiveCodeBench: 75.8IFEval: 83.9IFBench: 37.0MMLU: 85.3PopQA: 33.1GPQA: 55.7AlpacaEval: 69.1

Best scores per benchmark

benchmarkcategorybest score
MATHmath95.6
HumanEvalcoding90.0
BBHreasoning88.8
AGIEvalreasoning85.9
MMLUknowledge85.3
IFEvalinstruction-following83.9
LiveCodeBenchcoding75.8
AIME 2024math73.5
AlpacaEvalchat69.1
MBPPcoding66.7
AIME 2025math66.2
GPQAreasoning55.7
IFBenchinstruction-following37.0
PopQAknowledge33.1

Variants (1)

variantkindquantbitsvRAMdlorgyear
Olmo-3-32B-Think-DPOoriginalbf1616.0600allenai2026
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.