HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/allenai/allenai/Llama-3.1-Tulu-3-8B-DPO

allenai/Llama-3.1-Tulu-3-8B-DPO

family llama · 3 variants · unknown

provider
allenai
released
—
huggingface
https://huggingface.co/allenai/Llama-3.1-Tulu-3-8B-DPO
best scores
MMLU: 84.4PopQA: 55.7TruthfulQA: 46.8BBH: 88.0DROP: 61.3MATH: 63.4GSM8K: 76.2HumanEval: 86.2IFEval: 82.55AlpacaEval: 12.4GPQA: 6.26MUSR: 10.52MMLU-Pro: 20.23

Best scores per benchmark

benchmarkcategorybest score
BBHreasoning88.0
HumanEvalcoding86.2
MMLUknowledge84.4
IFEvalinstruction-following82.55
GSM8Kmath76.2
MATHmath63.4
DROPreasoning61.3
PopQAknowledge55.7
TruthfulQAsafety46.8
MMLU-Proknowledge20.23
AlpacaEvalchat12.4
MUSRreasoning10.52
GPQAreasoning6.26

Variants (3)

variantkindquantbitsvRAMdlorgyear
Llama-3.1-Tulu-3-8Boriginalbf1616.07,002allenai2025
Llama-3.1-Tulu-3.1-8Boriginalbf1616.0646allenai2025
Llama-3.1-Tulu-3-8B-DPO-GGUFggufgguf2.0324bartowski2024
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.