DeepSeek-R1 GGUF size and VRAM requirements

License: mit ⬇ 28,352 ❤ 1114
Parameters671.03B
Context163,840

unsloth/DeepSeek-R1-GGUF is a very large reasoning-focused model with 671.03 billion parameters, built on the deepseek2 architecture. It is released under the mit license and has been downloaded 28,352 times.

To run unsloth/DeepSeek-R1-GGUF locally at a 4,096-token context, its quantized versions need between 138.07 GB (IQ1_S, lowest quality) and 1257.56 GB (BF16, highest quality) of memory, weights plus KV cache and a system margin included.

→ Guide: How much VRAM do you need?

GGUF file size and memory by quantization

Compare real GGUF weight sizes, estimated KV cache and total memory for Q4, Q5, Q8 and every quantization published in this repository.

Quant.Bits QualityWeights KVTotal Speed~Verdict
IQ1_S 1.67 Very low 130.6 GB 6.67 GB 138.07 GB Insufficient
IQ1_M 2.01 Very low 157.32 GB 6.67 GB 164.79 GB Insufficient
IQ2_XXS 2.34 Very low 182.69 GB 6.67 GB 190.16 GB Insufficient
Q2_K_XS 2.64 Low 206.06 GB 6.67 GB 213.53 GB Insufficient
Q2_K_XL 2.7 Low 211.04 GB 6.67 GB 218.51 GB Insufficient
Q2_K 2.91 Low 227.27 GB 6.67 GB 234.74 GB Insufficient
Q2_K_L 2.91 Low 227.47 GB 6.67 GB 234.94 GB Insufficient
Q3_K_M 3.81 Fair 297.28 GB 6.67 GB 304.75 GB Insufficient
Q4_K_M 4.82 Good 376.65 GB 6.67 GB 384.13 GB Insufficient
Q5_K_M 5.67 Very good 442.75 GB 6.67 GB 450.22 GB Insufficient
Q6_K 6.57 Excellent 512.97 GB 6.67 GB 520.44 GB Insufficient
Q8_0 8.5 Excellent 664.3 GB 6.67 GB 671.77 GB Insufficient
BF16 16.0 Excellent 1250.09 GB 6.67 GB 1257.56 GB Insufficient

KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.

Frequently asked questions

Can I run unsloth/DeepSeek-R1-GGUF on an 8 GB GPU?

No. unsloth/DeepSeek-R1-GGUF does not fit on an 8 GB GPU, even with the smallest quantization and system RAM offloading.

Can I run unsloth/DeepSeek-R1-GGUF on a 16 GB GPU?

No. unsloth/DeepSeek-R1-GGUF does not fit on a 16 GB GPU, even with the smallest quantization and system RAM offloading.

Can I run unsloth/DeepSeek-R1-GGUF on a 24 GB GPU?

No. unsloth/DeepSeek-R1-GGUF does not fit on a 24 GB GPU, even with the smallest quantization and system RAM offloading.

What is the best quantization for unsloth/DeepSeek-R1-GGUF?

If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.