GGUF models by AmpereComputing

68 GGUF models quantized by AmpereComputing on Hugging Face. Sizes run from 494M to 671B parameters. The longest trained context is 10240K tokens.

Most downloadedNewestLargestLongest contextSmallest
GE
Gemma 3 1b It
Gemma · AmpereComputing
Params1B
Context32K
Smallest1.0 GB
GGUF
DE
DeepSeek R1 Distill Qwen 7B
DeepSeek · AmpereComputing
Params7.62B
Context128K
Smallest7.7 GB
ReasoningTool use
QW
Qwen3 4B
Qwen · AmpereComputing
Params4.02B
Context40K
Smallest4.1 GB
ReasoningTool use
QW
qwen 3 a3b 30b instruct
Qwen · AmpereComputing
Params30.5B
Context256K
Smallest31.0 GB
ReasoningTool use
DE
Deepseek-V3.1
DeepSeek · AmpereComputing
Params671B
Context160K
Smallest676.6 GB
ReasoningTool use
LL
Llama 3.1 8B Instruct
Llama · AmpereComputing
Params8.03B
Context128K
Smallest8.1 GB
Tool use
QW
Qwen2.5 3B Instruct
Qwen · AmpereComputing
Params3.09B
Context32K
Smallest3.1 GB
Tool use
QW
qwen 3 coder a3b 30b instruct
Qwen · AmpereComputing
Params30.5B
Context256K
Smallest31.0 GB
CodingTool use
GR
Granite 4.0 H Small
Granite · AmpereComputing
Params32.2B
Context1024K
Smallest32.7 GB
Tool use
QW
qwen 3 4b 2507 instruct
Qwen · AmpereComputing
Params4.02B
Context256K
Smallest4.1 GB
ReasoningTool use
GR
granite-8b-code-instruct-128k
Granite · AmpereComputing
Params8.05B
Context125K
Smallest8.1 GB
Coding
QW
Qwen3 8B
Qwen · AmpereComputing
Params8.19B
Context40K
Smallest8.3 GB
ReasoningTool use
QW
Qwen3 1.7B
Qwen · AmpereComputing
Params2.03B
Context40K
Smallest2.1 GB
ReasoningTool use
DE
DeepSeek R1 Distill Llama 70B
DeepSeek · AmpereComputing
Params70.6B
Context128K
Smallest71.1 GB
ReasoningTool use
QW
Qwen2.5 0.5B Instruct
Qwen · AmpereComputing
Params494M
Context32K
Smallest0.5 GB
Tool use
RA
RakutenAI 7B Chat
Rakutenai · AmpereComputing
Params7.37B
Context32K
Smallest7.4 GB
GGUF
MI
mistral small 3.2 2506 24b instruct
Mistral · AmpereComputing
Params23.6B
Context128K
Smallest23.8 GB
Tool use
GR
Granite 4.0 Tiny Preview
Granite · AmpereComputing
Params6.67B
Context1024K
Smallest6.7 GB
ReasoningTool use
QW
Qwen2.5 7B Instruct
Qwen · AmpereComputing
Params7.62B
Context32K
Smallest7.7 GB
Tool use
QW
Qwen2.5 32B Instruct
Qwen · AmpereComputing
Params32.8B
Context32K
Smallest33.0 GB
Tool use
QW
qwen 3 4b 2507 thinking
Qwen · AmpereComputing
Params4.02B
Context256K
Smallest4.1 GB
ReasoningTool use
QW
qwen 3 a3b 30b thinking
Qwen · AmpereComputing
Params30.5B
Context256K
Smallest31.0 GB
ReasoningTool use
DE
DeepSeek V3 0324 BF16
DeepSeek · AmpereComputing
Params671B
Context160K
Smallest676.6 GB
Tool use
BI
bielik v3.0 4.5b instruct
Bielik · AmpereComputing
Params4.76B
Context32K
Smallest4.8 GB
GGUF
QW
Qwen3 0.6B
Qwen · AmpereComputing
Params752M
Context40K
Smallest0.8 GB
ReasoningTool use
DE
DeepSeek R1 0528 Qwen3 8B
DeepSeek · AmpereComputing
Params8.19B
Context128K
Smallest8.3 GB
ReasoningTool use
DE
DeepSeek R1 Distill Llama 8B
DeepSeek · AmpereComputing
Params8.03B
Context128K
Smallest8.1 GB
ReasoningTool use
DE
DeepSeek R1 Distill Qwen 32B
DeepSeek · AmpereComputing
Params32.8B
Context128K
Smallest33.0 GB
ReasoningTool use
DE
DeepSeek R1 Distill Qwen 1.5B
DeepSeek · AmpereComputing
Params1.78B
Context128K
Smallest1.8 GB
ReasoningTool use
GR
Granite 3.3 2b Instruct
Granite · AmpereComputing
Params2.53B
Context128K
Smallest2.6 GB
ReasoningTool use

Keep browsing