GGUF models by Biscotto58

35 GGUF models quantized by Biscotto58 on Hugging Face. Sizes run from 3.21B to 14.8B parameters. The longest trained context is 1000K tokens.

Most downloadedNewestLargestLongest contextSmallest
MI
MistralNemoDionysusV3
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
QW
Qwen3 4B Instruct Ranker
Qwen · Biscotto58
Params4.02B
Context256K
Q4_K_M2.5 GB
ReasoningTool use
QW
Qwen3 14B Malice unhinged depraved
Qwen · Biscotto58
Params14.8B
Context40K
Q4_K_M9.0 GB
ReasoningTool use
MI
MistralNemoMegaV1 rev
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
GGUF
LL
Llama 3.2 3B Instruct WriterV2 4E
Llama · Biscotto58
Params3.21B
Context128K
Q4_K_M2.0 GB
Tool use
QW
Qwen3 4B Instruct WriterV2 6E
Qwen · Biscotto58
Params4.02B
Context256K
Q4_K_M2.5 GB
ReasoningTool use
MI
MistralNemoDionysus ID
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysusV2
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul Celeste DT
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul Celeste
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul Celeste DT V3
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul Magnum DT
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul Magnum Slerp MB
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus ID V2 magnlerp 4
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul Celeste DT V2
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
QW
Qwen3 4B Instruct Ranker 5000steps
Qwen · Biscotto58
Params4.02B
Context256K
Q4_K_M2.5 GB
ReasoningTool use
MI
MistralNemoDionysusV1
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
GGUF
MI
MistralNemoDionysus ID V2 faslerp
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
CA
CaptainNemo
Captainnemo · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Brain
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoMegaV0
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
GGUF
MI
MistralNemoMegaV1
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
GGUF
MI
MistralNemoDionysusV1 slerp
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
GGUF
MI
MistralNemoDionysus Soul Celeste Slerp
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul Magnum Slerp AEB
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus Soul Magnum Slerp CB
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus ID V2 SLERP
Mistral · Biscotto58
Params12.2B
Context1000K
Q4_K_M7.5 GB
Tool use
MI
MistralNemoDionysus ID V2 magnlerp
Mistral · Biscotto58
Params12.2B
Context128K
Q4_K_M7.5 GB
Tool use
QW
Qwen3 14B Malice
Qwen · Biscotto58
Params14.8B
Context40K
Q4_K_M9.0 GB
ReasoningTool use

About Biscotto58 on Hugging Face

Biscotto58 has published 35 GGUF models that ggufmodels.com tracks, across families such as Mistral, Qwen and Llama. The most downloaded are MistralNemoDionysusV3, Qwen3 4B Instruct Ranker and Qwen3 14B Malice unhinged depraved.

The quants Biscotto58 publishes most often are Q4_K_M (35). Their Q4_K_M files range from 2.0 to 9.0 GB. Parameter counts run from 3.21B to 14.8B, and the longest trained context window is 1000K tokens.

Frequently asked questions

What GGUF models has Biscotto58 published?

35 models are listed here, mainly from the Mistral, Qwen and Llama families. The most downloaded is MistralNemoDionysusV3.

Which quants does Biscotto58 publish?

The most common are Q4_K_M (35 models).

Keep browsing