Google unveils Gemini 4 Argon, a new AI model aimed at top-tier performance
Google has launched its Gemini 4 Argon model, touting superior reasoning and lower costs compared with leading competitors.
After a period of lagging behind rivals, Google introduced Gemini 4 Argon, claiming it delivers advanced reasoning and better performance across a range of demanding workflows. The company highlights improvements in reducing cliché responses and repetition, and asserts that Argon outperforms models such as GPT-6 Astra, Claude Fable 5.1 and Claude Opus 5.5 on benchmarks covering coding, cybersecurity, scientific research, finance and legal analysis.
Internally, Argon is already being used to locate and fix security vulnerabilities, cut the compute needed for YouTube comments by half, and debug the physics simulator for Alphabet's drone delivery unit Wing. The model rolls out today to higher-tier Pro and Ultra subscribers via the Gemini app and the Antigravity development platform, with a launch price of $2 per million input tokens and $10 per million output tokens—about one-fifth the cost of comparable frontier models.
Pricing will double after the introductory period ends in late November, but Argon is expected to remain cheaper on a per-token basis. The release follows a limited test with trusted cyber defenders and arrives shortly after Meta announced its own frontier model, Muse Wonder Full.
How this was covered
- Coverage peaked at 6 outlets in a single hour
Why it matters
Google's Argon launch signals a renewed push in the competitive AI market, offering cheaper, high-performance options for developers and enterprises.
In this story
