gemma-3-4b-sft-persona-gguf-Q8_0 : GGUF

• For text only LLMs: ./llama.cpp/llama-cli -hf shemilk/gemma-3-4b-sft-persona-gguf-Q8_0 --jinja
• For multimodal models: ./llama.cpp/llama-mtmd-cli -hf shemilk/gemma-3-4b-sft-persona-gguf-Q8_0 --jinja

Available Model files:

• gemma-3-4b-sft-ml-merged.Q8_0.gguf
• gemma-3-4b-sft-ml-merged.F16-mmproj.gguf

Note

The model's BOS token behavior was adjusted for GGUF compatibility. This was trained 2x faster with Unsloth