BC-250 Linux Box Gets a Local AI Agent Stack With Qwen 3.5
A new build script kit brings llama.cpp and Qwen 3.5-9B to the AMD BC-250 mining board running CachyOS.
Developer nielsek published bc250-llama.cpp, a script collection that builds and runs llama.cpp with Vulkan on the AMD BC-250, an ex-mining board packing a cut-down PS5 APU, known in the Linux scene as Cyan Skillfish. It runs CachyOS.
The setup serves unsloth/Qwen3.5-9B-GGUF:Q6_K headless via llama-server on port 8000 for coding tools like opencode. Because the BC-250 shares 16 GB of GDDR6 between CPU and GPU, the scripts rely on the ttm.pages_limit=3145728 kernel parameter from the bc250-toolkit to let the graphics driver claim up to 12 GiB of system memory. The setup assumes a fully unlocked BC-250 with all CPU cores and GPU CUs enabled.
The repository has four shell scripts:
./build.shinstalls dependencies, clones, and builds llama.cpp../test.shverifies GPU detection and checks that the model runs fast enough../run.shlaunchesllama-serverinside a tmux session and leaves the desktop to free about 3 GB of RAM../stop.shshuts down the server and returns to the desktop.
The 9B Qwen model fits the 12 GiB GPU allowance with a 128K context. On this hardware it reads prompts at roughly 400 tokens per second and generates at about 40 tokens per second.