Vibeleaderboard
← All Intel
Intel / video

Generative Video at the Speed of Light — Keegan McCallum, uRun

Source
youtube.com
Author
AI Engineer
Date

Context

uRun's video-serving talk describes an interaction in which people steer generated imagery continuously, including a proposed webcam preview of a different haircut. Making that useful requires user controls and generated frames to stay synchronized. The serving system must coordinate several models in one stream and place computing resources near users. The talk's serving discussion.

This changes the engineering objective from delivering a finished clip to maintaining a responsive session. A model that generates images quickly is one component; the network and pipeline must carry those updates back while incorporating the person's next input. The talk presents this coordination as the remaining serving challenge. Interaction and infrastructure account.

Terms in this piece · Glossary
  • streaming — Sending a model's response token by token as it is generated, so the reader sees text immediately instead of waiting for the whole answer.
  • distillation — Training a small, cheap model to imitate a big one's outputs, keeping much of the capability at a fraction of the cost.
Read the source www.youtube.com
More from AI Engineer
Recommended reads
Comments

Checking sign-in…

Loading comments…