search

szczyglis-dev / py-gpt

Desktop AI Assistant powered by GPT-4, GPT-4 Vision, GPT-3.5, Gemini, Claude, Llama 3, DALL-E, Langchain, Llama-index, chat, vision, voice control, image generation and analysis, agents, code/command execution, file upload/download, speech synthesis and recognition, access to Web, memory, presets, plugins, assistants, and more. Linux, Windows, Mac.
https://pygpt.net
MIT License
544 stars 106 forks source link

feature request - paste image #44

Closed glinkot closed 5 months ago

glinkot commented 5 months ago

For speed of workflow, I'd love it if I could ctrl-V when entering my input, to add the current clipboard image (or other media type I guess) as an attachment to my GPT Vision prompt or similar.

The workflow of taking a screenshot, saving it to a folder, then browsing to it from pyGPT adds some friction when doing stuff like "here is my interface, what should I click on' type things.

szczyglis-dev commented 5 months ago

Pasting images as attachments in input using ctrl-V added in 2.1.45. Thanks!

By the way, you can take screenshots using the 'Ask with screenshot' option from the tray icon - an attachment is created automatically.

glinkot commented 5 months ago

Fantastic! Thanks so much mate!