HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/allenai/allenai/Olmo-3.1-32B-Instruct-DPO

allenai/Olmo-3.1-32B-Instruct-DPO

family olmo · 1 variants · unknown

provider
allenai
released
—
huggingface
https://huggingface.co/allenai/Olmo-3.1-32B-Instruct-DPO
best scores
MATH: 74.4AIME 2024: 12.7AIME 2025: 8.2BBH: 69.0AGIEval: 71.7HumanEval: 80.8MBPP: 61.5LiveCodeBench: 35.4IFEval: 87.7IFBench: 29.7MMLU: 79.0PopQA: 23.7GPQA: 41.3AlpacaEval: 42.2

Best scores per benchmark

benchmarkcategorybest score
IFEvalinstruction-following87.7
HumanEvalcoding80.8
MMLUknowledge79.0
MATHmath74.4
AGIEvalreasoning71.7
BBHreasoning69.0
MBPPcoding61.5
AlpacaEvalchat42.2
GPQAreasoning41.3
LiveCodeBenchcoding35.4
IFBenchinstruction-following29.7
PopQAknowledge23.7
AIME 2024math12.7
AIME 2025math8.2

Variants (1)

variantkindquantbitsvRAMdlorgyear
Olmo-3.1-32B-Instructoriginalbf1616.050,114allenai2026
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.