Skip to content
Haydn Belfield· @HaydnBelfield · X·Indexed by this site · Originally published· AI updates →· Its place in the scan record →

Google releases Gemini 4 Neon: focused on fast task completion

AI summary

Google today launched Gemini 4 Neon, saying it has frontier-level intelligence and can quickly complete end-to-end tasks in complex software engineering, knowledge work and scientific workflows. According to the official announcement, Neon has greatly improved steerability, ranks second on Agent Arena Steering, has strong multi-step instruction following, and has the lowest confabulation rate to date in Google Antigravity internal tests; it ranks third on the Artificial Analysis Intelligence Index (61 points in the xhigh tier), behind only larger and slower models, and ranks second on the Vals Index, on the Pareto frontier. The news was announced by koraykv on X and reposted by Haydn Belfield.

Full text · Original

New model dropped ->

Quoted postkoray kavukcuoglu@koraykv
Today we’re introducing Gemini 4 Neon — bringing frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work, and science workflows. We focused on making Neon better to work with: much improved steerability (#2 on Agent Arena Steering), strong multi-step instruction following, and our lowest fabrication rate yet in internal Google @Antigravity testing. Neon ranks #3 (61 at xhigh) on the Artificial Analysis Intelligence Index — landing just behind larger, slower models — and #2 on the Vals Index on the Pareto frontier.
View the quoted post on X

Source: Haydn Belfield · x.com