Description
Fine-tuning and deploying Google's Gemma on your own servers can be hard to start. This is Google DeepMind's official Gemma open-weight LLM library with tools for inference, fine-tuning and sampling.
Built on JAX, it supports multimodal input, LoRA and multi-turn chat, with docs and examples.
Official:From DeepMind.
Fine-tuning:LoRA.
Multimodal:Text and images.
JAX:Efficient.
Built on JAX, it supports multimodal input, LoRA and multi-turn chat, with docs and examples.
Features
Official:From DeepMind.
Fine-tuning:LoRA.
Multimodal:Text and images.
JAX:Efficient.
Screenshots
Tags:llm