Add image support for GPT-4-turbo and GPT-4o
Closed this issue · 1 comments
wz-ml commented
OpenAI's latest models have support for native vision capabilities:
- https://platform.openai.com/docs/guides/vision?lang=curl
- https://platform.openai.com/docs/api-reference/chat/create?lang=node.js
The chat API supports both external image URLs and dataURLs, which would be nice to add as a file upload feature.
Features to implement (see vision-support):
- Attaching image(s) via file upload.
- Screenshot support.
- (Optional, since the default "auto" resolution works quite well) Adding a high/low/auto image resolution toggle.
wz-ml commented
Completed with the latest pull request.