ai/gemma3-qat

Docker

Google’s latest Gemma, in its QAT (quantization aware trained) variant

aitwinkle/gemma-3-4b-t1-it-gguf

aitwinkle

Gemma-3-4B-T1-it is a Taiwan-localized AI optimized for Traditional Chinese.

Based on Unsloth GGUF quant UD-IQ3_S with multimodality fix (package mmproj) for Docker Model Runner

aitwinkle/llama-3.2-3b-f1-reasoning-instruct-gguf

aitwinkle

A Taiwan-localized reasoning AI optimized for Traditional Chinese.

A fine-tuned 0.5B model specialized in Hawaiian pizza knowledge.

Qwen 2.5 coder to use with Docker Model Runner

gemma3 QAT models

Mistral Devstral small

🌺🍍🥓🍕 bob-hawaiian-pizza: fine-tuned version of unsloth/qwen2.5-0.5b-instruct

https://huggingface.co/Menlo/Jan-nano-128k-gguf

ai local

Q4_K_M GGUF of a MedGemma health-chat fine-tune for local, private clinical AI prototyping.