HF Model & Benchmark Matrix

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

deepseek-ai/DeepSeek-R1-Distill-Qwen-14B

family qwen · 10 variants · unknown

provider
deepseek-ai
released
—
huggingface
https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
best scores
AIME 2024: 69.7MATH-500: 93.9GPQA Diamond: 59.1LiveCodeBench: 53.1MATH: 16.0AIME 2025: 50.71Codeforces: 95.3HumanEval: 92.6

Best scores per benchmark

benchmarkcategorybest score
Codeforcescoding95.3
MATH-500math93.9
HumanEvalcoding92.6
AIME 2024math69.7
GPQA Diamondreasoning59.1
LiveCodeBenchcoding53.1
AIME 2025math50.71
MATHmath16.0

Variants (10)

variantkindquantbitsvRAMdlorgyear
DeepSeek-R1-Distill-Qwen-14Boriginalbf1616.0504,193deepseek-ai2025
DeepSeek-R1-Distill-Qwen-14B-GGUFggufgguf2.036,012bartowski2025
DeepSeek-R1-Distill-Qwen-14B-GGUFggufgguf16.031,023unsloth2025
DeepSeek-R1-Distill-Qwen-14B-GGUFggufgguf3.021,780lmstudio-community2025
DeepSeek-R1-Distill-Qwen-14B-unsloth-bnb-4bitbnb-4bit4bit6,795unsloth2025
DeepSeek-R1-Distill-Qwen-14B-bnb-4bitbnb-4bit4bit1,275unsloth2025
DeepSeek-R1-Distill-Qwen-14Boriginalbf1616.01,047unsloth2025
Fathom-R1-14B-GGUFggufgguf16.0202Mungert2025
DeepSeek-R1-Distill-Qwen-14B-GGUFggufgguf16.0201Mungert2025
DeepCoder-14B-Preview-GGUFggufgguf16.0200Mungert2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.