WHAT NEW IN X10.30 FRI

AI collections from X · 22 collections · 60,637 posts

gemini-4POST

Gemini 4 benchmarks: knowledge SOTA, coding around Astra/Fable 5.1 level

A post says Gemini 4 is state of the art on knowledge benchmarks, with coding performance roughly on par with Astra/Fable 5.1, and worries that safety review could delay launch.

Haider.
@haider1Archived 27 days agoOriginally posted 2026.10.01
Haider. @haider1Originally posted 2026.10.01
Open original on X

so we've got Gemini 4 as for benchmarks, it's SOTA in knowledge, while coding performance seems to be around Astra/Fable 5.1 level NOT BAD but now it comes down to how fast they can get through the safety rollout, so it doesn't become obsolete by the time Fable 5.5 and 6.1 Astra launch

x.com/haider1/status/2105432512566796385Archived 2026.10.02 10:14