5 ms·
LLM + Text-to-Image model is exactly how DALL·E 3 is deployed, fwiw
by alextheparrot 3y ago
LLM + Text-to-Image model is exactly how DALL·E 3 is deployed, fwiw
- zaptrem 3y agoIncluding the text positioning generation part? What’s the source on that?
- alextheparrot 3y agoThe comment was directed at “doesn't this method add another cost and overhead for calling Text-to-Image models”
- wahnfrieden 3y agoNo