| cd llama.cpp_omega | |
| cmake -B build -DGGML_VULKAN=ON && cmake --build build -j | |
| echo "/llama.cpp_omega/build/bin" | sudo tee /etc/ld.so.conf.d/llama-custom.conf | |
| sudo ldconfig | |
| if error ## wget -O vendor/miniaudio/miniaudio.h https://raw.githubusercontent.com/mackron/miniaudio/master/miniaudio.h | |
| #server | |
| "/home/userk21/llama.cpp_offellia/build/bin/llama-server" \ | |
| -m "/home/GGUFS/ΩFFΣLLIα_Q4_2_H_gemma-4-12B-it.gguf" \ | |
| -md "/home/GGUFS/ΩFFΣLLIα_Q4_2_H_gemma-4-12B-it-assistant.gguf" \ | |
| -c 50000 \ | |
| -ngl 42 --n-cpu-moe 42 -fa 1 \ | |
| -np 1 \ | |
| -fa on \ | |
| --parallel 1 \ | |
| --spec-type draft-mtp --spec-draft-n-max 4 --spec-draft-n-min 1 --spec-draft-p-min 0.75 \ | |
| --dry-multiplier 0.8 --dry-base 1.75 --dry-allowed-length 2 \ | |
| --agent \ | |
| --tools all \ | |
| --reasoning auto \ | |
| --swa-full \ | |
| --kv-unified \ | |
| --webui-mcp-proxy \ | |
| --threads-http -1 \ | |
| --port 5173 \ | |
| --host 127.0.0.1 | |