News for hackers. Stuff that matters.

A comment on DeepSeek open-sources inference optimizations with 60–85% faster generation [pdf]. Parent comment

Re:DeepSeek open-sources inference optimizations with 60–85%... by nozzlegear on Saturday June 27, 2026 @11:08AM (#48698930)
No more revealing than OpenAI, Anthropic and Google always having some new model that just so happens to be waiting in the wings whenever their competitors announce their own model bump.
[ Parent | on HN ]
Re:DeepSeek open-sources inference optimizations with 60–85%... by mycall on Sunday June 28, 2026 @12:20AM (#48704308)
That's because OpenAI, Anthropic and Google work on many models in parallel which work cooperatively from the user's POV. So GPT-5.6 is just a checkpoint of their multi-model development.
[ Parent | on HN ]