桌面端设置模型的时候需要增加一个输入类型的设置 我设置的模型支持图片输入，但是无法直接设置 需要改配置文件 #54995

Activity

1. opencode-agent commented on Oct 13, 2026

   

   opencode-agentBot

   on Oct 13, 2026 – with OpenCode Agent

   Contributor

   More actions

   Thanks for the suggestion! As we understand it, you'd like the desktop app's custom provider/model setup to have an input type setting (for example, "supports image input"), so you don't have to edit the config file to mark a model as vision-capable. Right now the form only asks for each model's ID and name. Maintainers will consider this.

   Related issues:

   • Images pasted with a custom-provider model are silently replaced by a "[Attached image/png: ...]" text placeholder — no warning, even when the upstream model has vision #47480: images sent to custom-provider models being replaced with a text placeholder
   • docs(providers): document attachment/modalities model capabilities for custom models #47482: docs for setting attachment/modalities capabilities on custom models in the config (the current workaround)
   • [V2 Desktop] Pasted images show no thumbnail when the active model is text-only #54991: no image thumbnails in the desktop app when the model is text-only
   • desktop: custom provider models always show reasoning badge as not supported despite capabilities.output #50608: the desktop app's reasoning badge for custom provider models