Adding Ctrl+v for vision models on Windows

continuedev / continue

⏩ Continue is the leading open-source AI code assistant. You can connect any models and any context to build custom autocomplete and chat experiences inside VS Code and JetBrains

https://docs.continue.dev/

Apache License 2.0

14.91k stars 1.1k forks source link

Adding Ctrl+v for vision models on Windows #1033

Open prefonty opened 5 months ago

prefonty commented 5 months ago

Validations

[ ] I believe this is a way to improve. I'll try to join the Continue Discord for questions
[X] I'm not able to find an open issue that requests the same enhancement

Problem

Regarding the Ctrl+v with Vision models (GPT-4v, Claude 3 Opus) on windows 11 machines, I try to use it in Cursor, VSCode, PhpStorm all the same, need to upload by clicking.

Solution

Having it work with a direct Ctrl+v right there in the chatbox would be much quicker.

sestinj commented 5 months ago

Just adding some information here as context: it appears that this is a problem specifically on Windows, and it might be related to https://github.com/ueberdosis/tiptap/issues/2912

ypref commented 5 months ago

Let's keep in mind that the feature works in Cursor with GPT-4v. The tab right next to .continue. So to me, we should not look into a pure system issue. Cursor devs found a workaround. I will keep digging for an open source project which found the solution so we can perhaps find some inspiration.

prefonty commented 5 months ago

Cursor also now supports Claude 3 Opus directly, with my own API key. And with it too the pasting of images from the clipboard works directly in the chatbox.