Hugging Face's logo

Barrrrry
/
DeepSeek-R1-W4AFP8

Safetensors
deepseek_v3
custom_code
8-bit precision
Model card Files
xet
Community
3

DeepSeek-R1-W4AFP8

This model is a mixed-precision quantized DeepSeek-R1, with dense layer using FP8_BLOCK_SCALING, MoE layers uses INT4 weights and FP8 activation.

Downloads last month
1,596
Safetensors
Model size
357B params
Tensor type
F32
·
BF16
·
F8_E4M3
·
I8
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Barrrrry/DeepSeek-R1-W4AFP8

Base model

deepseek-ai/DeepSeek-R1
Quantized
(68)
this model