HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/deepseek-ai/deepseek-ai/DeepSeek-R1-Distill-Llama-8B

deepseek-ai/DeepSeek-R1-Distill-Llama-8B

family llama · 8 variants · unknown

provider
deepseek-ai
released
—
huggingface
https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B
best scores
AIME 2024: 50.4MATH-500: 89.1GPQA Diamond: 49.0LiveCodeBench: 39.6MATH: 16.0

Best scores per benchmark

benchmarkcategorybest score
MATH-500math89.1
AIME 2024math50.4
GPQA Diamondreasoning49.0
LiveCodeBenchcoding39.6
MATHmath16.0

Variants (8)

variantkindquantbitsvRAMdlorgyear
DeepSeek-R1-Distill-Llama-8Boriginalbf1616.0408,725deepseek-ai2025
DeepSeek-R1-Distill-Llama-8B-GGUFggufgguf16.031,585unsloth2025
DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bitbnb-4bit4bit11,377unsloth2025
DeepSeek-R1-Distill-Llama-8B-GGUFggufgguf2.09,333bartowski2025
DeepSeek-R1-Distill-Llama-8B-GGUFggufgguf3.07,009lmstudio-community2025
DeepSeek-R1-Distill-Llama-8Boriginalbf1616.02,155unsloth2025
DeepSeek-R1-Distill-Llama-8B-bnb-4bitbnb-4bit4bit1,534unsloth2025
DeepSeek-R1-Distill-Llama-8B-GGUFggufgguf16.0126Mungert2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.