70k32

joined 1 year ago
[โ€“] 70k32@sh.itjust.works 1 points 4 months ago

jan.ai with some local model

[โ€“] 70k32@sh.itjust.works 1 points 4 months ago

This. Llama.cpp with Vulkan backend running in docker-compose, some Qwen3-Coder quantization from huggingface and pointing Opencode to that local setup with a OpenAI-compatible is working great for me.