AmpereComputing/deepseek-r1-0528-qwen-3-8b-gguf
8B
•
Updated
•
4
DeepSeek R1
Ampere's quantization formats (Q4_K_4 / Q8R16) require Ampere optimized llama.cpp available here: https://hub.docker.com/r/amperecomputingai/llama.cpp
-
-
AmpereComputing/deepseek-r1-0528-gguf
671B • Updated • 40 -
AmpereComputing/deepseek-r1-distill-llama-8b-gguf
8B • Updated • 10 -
AmpereComputing/deepseek-r1-distill-llama-70b-gguf
71B • Updated • 12 -
AmpereComputing/deepseek-r1-distill-qwen-1.5b-gguf
2B • Updated • 8 -
AmpereComputing/deepseek-r1-distill-qwen-7b-gguf
8B • Updated • 12 -
AmpereComputing/deepseek-r1-distill-qwen-14b-gguf
15B • Updated • 2 -
AmpereComputing/deepseek-r1-distill-qwen-32b-gguf
33B • Updated • 33 -
AmpereComputing/deepseek-r1-gguf
671B • Updated • 59