Run a Gemma 2B Model

[Khalilbz]

The latest version of Llama.cpp supports Gemma, I have downloaded the 2B GGUF model and it didn't work, is there a config or something I have to do to make it work ?

oo@oo-workstation:~/xxxx/drive/DEV/c-ai$ npx --no node-llama-cpp chat --model models/gemma-2b-GGUF/gemma-2b.gguf 
npm WARN config init.author.name Use `--init-author-name` instead.
npm WARN config init.version Use `--init-version` instead.
gguf_init_from_file: invalid magic characters 'vers'
llama_model_load: error loading model: llama_model_loader: failed to load model from /home/oo/xxxx/drive/DEV/c-ai/models/gemma-2b-GGUF/gemma-2b.gguf

llama_load_model_from_file: failed to load model
Error: Failed to load model
    at new LlamaModel (file:///home/oo/xxxx/drive/DEV/c-ai/node_modules/node-llama-cpp/dist/llamaEvaluator/LlamaModel.js:50:23)
    at RunChat (file:///home/oo/xxxx/drive/DEV/c-ai/node_modules/node-llama-cpp/dist/cli/commands/ChatCommand.js:187:19)
    at async Object.handler (file:///home/oo/xxxx/drive/DEV/c-ai/node_modules/node-llama-cpp/dist/cli/commands/ChatCommand.js:168:13)

[giladgd]

Version 2 doesn't support Gemma, you can use version 3 beta instead for that