Open atrokhym opened 3 days ago
Could you please post a video of how the interface works? This is a major feature involving vision, so it may have to be prioritized against the roadmap in the coming days. Thanks for the submission!
We also have many models and providers, probably we need to disable such features when untested provider/model are used
We also have many models and providers, probably we need to disable such features when untested provider/model are used
The upload button could be only available for those models with vision capabilities and frozen/disabled for the others. The dynamic LLMs can be fetched from the OpenRouter API and filter those which have "modality": "text+image->text"
. The ones that are hardcoded can be flagged in the constants.ts
@atrokhym Thank you so much for this, I hope we can get it merged in soon! I appreciate your reviews a ton too @pjmartorell, fantastic!
@pjmartorell thank you for doing my PR review. i think i incorporated changes you requested. pls take a look thanks a lot!
@atrokhym Thank you! Let me know if you can get around to reviewing again @pjmartorell!
@atrokhym Thank you! Let me know if you can get around to reviewing again @pjmartorell! @coleam00 -i did run though those issues @pjmartorell noted. i implemented all of them and ping him to review again. there are few commits after that. also - should mention - i might missing something about process like who do what, notifications, etc... so pls correct or point me in right direction. this is my first git hub public PR :-)