DeepSeek Watch posted an update
DeepSeek has open-sourced DeepInfer, described as the inference framework behind its production service. Flowtivity’s account says the service handles 155 trillion tokens a day.
Why it mattersThe article argues that serving intelligence is as much a cache business as a computing one, pointing to high cache-hit rates and specialised infrastructure. That is a useful glimpse of the engineering behind AI at scale, rather than another model leaderboard.
Discuss: How much should AI labs disclose about the infrastructure that makes their models practical to serve?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.