HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/deepseek-ai/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B

deepseek-ai/DeepSeek-R1-Distill-Qwen-32B

family qwen · 11 variants · unknown

provider
deepseek-ai
released
—
huggingface
https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
best scores
AIME 2024: 79.8MATH-500: 94.3GPQA Diamond: 71.5LiveCodeBench: 65.9MATH: 16.0AIME 2025: 70.0

Best scores per benchmark

benchmarkcategorybest score
MATH-500math94.3
AIME 2024math79.8
GPQA Diamondreasoning71.5
AIME 2025math70.0
LiveCodeBenchcoding65.9
MATHmath16.0

Variants (11)

variantkindquantbitsvRAMdlorgyear
DeepSeek-R1-Distill-Qwen-32Boriginalbf1616.0832,252deepseek-ai2025
DeepSeek-R1-Distill-Qwen-32B-GGUFggufgguf2.029,132bartowski2025
DeepSeek-R1-Distill-Qwen-32B-GGUFggufgguf16.023,089unsloth2025
DeepSeek-R1-Distill-Qwen-32B-GGUFggufgguf3.015,283lmstudio-community2025
DeepSeek-R1-Distill-Qwen-32B-bnb-4bitbnb-4bit4bit1,612unsloth2025
DeepSeek-R1-Distill-Qwen-32Boriginalbf1616.0571unsloth2025
DeepSeek-R1-Distill-Qwen-32B-unsloth-bnb-4bitbnb-4bit4bit288unsloth2025
DeepSeek-R1-Distill-Qwen-32B-GGUFggufgguf16.0265Mungert2025
QwenLong-L1-32B-GGUFggufgguf16.0154Mungert2025
DeepSeek-R1-Distill-Qwen-32B-Q8_0-GGUFggufgguf8.0103ggml-org2025
FairyR1-32B-GGUFggufgguf16.012Mungert2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.