When I tried Claude Sonnet 4.5 via Straico's API endpoint v2, I got always this error message: "anthropic/claude-sonnet-4.5 cannot read image input" whereas that model has input image vision (tested in several AI Clients).
Does DOAI-Proxy support input image vision in OpenCode ?
When I tried Claude Sonnet 4.5 via Straico's API endpoint v2, I got always this error message: "anthropic/claude-sonnet-4.5 cannot read image input" whereas that model has input image vision (tested in several AI Clients).
Does DOAI-Proxy support input image vision in OpenCode ?