#!/bin/bash

export WORK="/home/user/app"
cd $WORK

echo "✅ Downloading llamacpp..."

wget -O llama_cpp.zip https://github.com/ggml-org/llama.cpp/releases/download/b6102/llama-b6102-bin-ubuntu-x64.zip > /dev/null 2>&1
unzip llama_cpp.zip > /dev/null 2>&1

echo "✅ Booting up llama server..."

wget -O model.gguf https://huggingface.co/lmstudio-community/Qwen3-4B-Thinking-2507-GGUF/resolve/main/Qwen3-4B-Thinking-2507-Q6_K.gguf?download=true > /dev/null 2>&1
# wget -O model.gguf https://huggingface.co/lmstudio-community/Qwen3-4B-Instruct-2507-GGUF/resolve/main/Qwen3-4B-Instruct-2507-Q6_K.gguf?download=true > /dev/null 2>&1
# wget -O model.gguf https://huggingface.co/unsloth/gpt-oss-20b-GGUF/resolve/main/gpt-oss-20b-F16.gguf?download=true > /dev/null 2>&1
./build/bin/llama-server -m model.gguf --port 8000 --host 0.0.0.0 --threads 2 --ctx-size 4096  --mlock --jinja

echo "✅ llama server running on port 8000"