HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/deepseek-ai/deepseek-ai/DeepSeek-R1-Distill-Llama-70B

deepseek-ai/DeepSeek-R1-Distill-Llama-70B

family llama · 9 variants · unknown

provider
deepseek-ai
released
—
huggingface
https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B
best scores
AIME 2024: 70.0MATH-500: 94.5GPQA Diamond: 65.2LiveCodeBench: 57.5MATH: 16.0

Best scores per benchmark

benchmarkcategorybest score
MATH-500math94.5
AIME 2024math70.0
GPQA Diamondreasoning65.2
LiveCodeBenchcoding57.5
MATHmath16.0

Variants (9)

variantkindquantbitsvRAMdlorgyear
DeepSeek-R1-Distill-Llama-70Boriginalbf1616.0778,263deepseek-ai2025
DeepSeek-R1-Distill-Llama-70B-GGUFggufgguf1.076,852bartowski2025
DeepSeek-R1-Distill-Llama-70B-GGUFggufgguf16.031,631unsloth2025
DeepSeek-R1-Distill-Llama-70B-GGUFggufgguf3.08,424lmstudio-community2025
DeepSeek-R1-Distill-Llama-70B-bnb-4bitbnb-4bit4bit2,203unsloth2025
DeepSeek-R1-Distill-Llama-70B-GGUFggufgguf16.0280Mungert2025
DeepSeek-R1-Distill-Llama-70B-unsloth-bnb-4bitbnb-4bit4bit66unsloth2025
DeepSeek-R1-Distill-Llama-70Boriginalbf1616.056unsloth2025
DS-R1-Distill-70B-ArliAI-RpR-v4-Largeoriginalbf1616.037ArliAI2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.