HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/Qwen/Qwen/Qwen2.5-7B

Qwen/Qwen2.5-7B

family qwen · 24 variants · unknown

provider
Qwen
released
—
huggingface
https://huggingface.co/Qwen/Qwen2.5-7B
best scores
GPQA: 31.6MMLU-Pro: 47.5Humanity's Last Exam: 11.8LiveCodeBench: 28.6SciCode: 100.0AIME 2024: 55.5AIME 2025: 45.6BBH: 62.7MATH-500: 72.8MATH: 81.8

Best scores per benchmark

benchmarkcategorybest score
SciCodecoding100.0
MATHmath81.8
MATH-500math72.8
BBHreasoning62.7
AIME 2024math55.5
MMLU-Proknowledge47.5
AIME 2025math45.6
GPQAreasoning31.6
LiveCodeBenchcoding28.6
Humanity's Last Examknowledge11.8

Variants (24)

variantkindquantbitsvRAMdlorgyear
Qwen2.5-7B-Instructoriginalbf1616.011,997,940Qwen2025
Qwen2.5-Coder-7Boriginalbf1616.01,037,293Qwen2024
Qwen2.5-7Boriginalbf1616.0884,975Qwen2024
Qwen2.5-Math-7Boriginalbf1616.0155,451Qwen2024
Qwen2.5-7B-Instruct-1Moriginalbf1616.058,258Qwen2025
Qwen2.5-7B-bnb-4bitbnb-4bit4bit10,800unsloth2025
Qwen2.5-7B-unsloth-bnb-4bitbnb-4bit4bit7,036unsloth2025
Qwen2.5-7Boriginalbf1616.05,574unsloth2025
OpenReasoning-Nemotron-7Boriginalbf1616.01,403nvidia2025
MolmoAct-7B-D-0812originalbf1616.0889allenai2025
Qwen2.5-7B-Instruct-1M-GGUFggufgguf16.0583Mungert2025
MolmoAct-7B-D-Pretrain-RT-1-0812originalbf1616.0391allenai2025
Qwen2.5-7B-Instruct-GGUFggufgguf16.0387Mungert2025
SWE-Fixer-Retriever-7Boriginalbf1616.0366internlm2025
SynLogic-7Boriginalbf1616.0364MiniMaxAI2025
MolmoAct-7B-D-Pretrain-0812originalbf1616.0329allenai2025
MolmoAct-7B-D-LIBERO-Object-0812originalbf1616.0220allenai2025
MolmoAct-7B-D-LIBERO-Spatial-0812originalbf1616.0190allenai2025
MolmoAct-7B-D-LIBERO-Long-0812originalbf1616.0142allenai2025
MolmoAct-7B-D-LIBERO-Goal-0812originalbf1616.0134allenai2025
MolmoAct-7B-D-Captioner-0812originalbf1616.053allenai2025
OREAL-7B-SFToriginalbf1616.052internlm2025
ASearcher-Local-7Boriginalbf1616.026inclusionAI2025
ASearcher-Web-7Boriginalbf1616.024inclusionAI2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.