Loaded instance reports image input; llama.cpp attaches mmproj-Q8_0 from the HF cache. Mark it vision in OpenCode and drop the old text-only comment.
Managed /etc/opencode/opencode.json points at straylight:11434 with the router model ids, Laguna as default, Ornith 1.5 9B as small_model, and a vision subagent on Qwen3-VL-8B abliterated.
kyle