Back to Quant Hub

DeepSeek-R1-Distill-Qwen-7B

7B

DeepSeek

R1 reasoning in a 7B footprint. Best value for 8–12GB VRAM CoT experiments.

⬇ 51.5K HF downloads♥ 131 likesbartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF· stats from 6/24/2026

Consumer GPUMac / Apple SiliconCPU / VPS

131K

Max Context

2

Quant Variants

EXL2 4.65bpw

Best Quality

97.8%

Accuracy Retained

Calculate VRAM Hugging Face Compare

Quantization Variants

Per-quant VRAM, quality loss, and inference speed on RTX 4090

Format	Level	BPW	VRAM	PPL Loss	Speed	Actions
GGUF	Q4_K_M	4.85	5.4 GB	3.0%	152 tok/s	Calc HF
EXL2	4.65bpw	4.65	5.2 GB	2.2%	210 tok/s	Calc HF