metadata
license: gemma
base_model:
- google/gemma-3-1b-it
- unsloth/gemma-3-1b-it
- unsloth/gemma-3-1b-it-GGUF
This is just a reupload of the quantized 4Q K M version of Gemma 3 1b by Google and unsloth. It is used for benchmark purposes on the Jetson Nano and llama.cpp with CUDA support:
- https://github.com/kreier/llama.cpp-jetson - 🖳 instructions to compile llama.cpp with CUDA support
- https://github.com/kreier/llama.cpp-jetson.nano - 🚀 precompiled versions that can be installed on the Jetson in minutes
- https://github.com/kreier/jetson a few other insights and results