Run it in the Llama app: https://llama.app
Or from the terminal:
llama serve -hf ggml-org/gemma-4-E4B-it-GGUF
[!IMPORTANT] This model is automatically converted using https://github.com/ggml-org/convert
google/gemma-4-E4B-it
LocalMaxxing/data/bt/os/llama.cpp/llama.cpp-bench-upstream/build-bench-cuda-gcc15/bin/llama-server -m /data/bt/models/hub/models--ggml-org--gemma-4-E4B-it-GGUF/snapshots/2714b5519c6c3516b1000e7c5e1eba998dfe1fe8/gemma-4-E4B-it-Q8_0.gguf -a gemma-4-e4b-it-q8_0 -ngl 999 -fa on -c 4096 -b 2048 -ub 512 -np 1 --no-context-shift --metrics --no-webui --host 127.0.0.1 --port 18106在 LocalMaxxing 上查看测试
cd /data/local/tmp/llmfit-llamacpp-android-vulkan && LD_LIBRARY_PATH=bin bin/llama-bench -m models/gemma-4-E4B-it-Q4_K_M.gguf -p 512 -n 128 -r 2 -o json -ngl 99 -fa 1 -t 8 -b 512 -ub 256 -dev Vulkan0在 LocalMaxxing 上查看测试
来自 LocalMaxxing 的 5 次社区测试（2 种配置）的中位数。
自己运行：配置指南