Google details Gemini 4 Argon access and pricing

Google announced Gemini 4 Argon as a new frontier model on September 30, 2026.[1] The company says it is built for deep reasoning across complex, long-horizon workflows and supports work in software engineering, enterprise knowledge work and cybersecurity defense.[1] Google also describes it as designed for coding, finance, legal work, cybersecurity and science.[2] That scope positions Argon for tasks that involve several stages of work, but the evidence does not specify model limits, technical specifications or independent evaluations beyond Google’s stated claims.[1][2]

Google dated its broader-availability announcement October 13, 2026, and said Argon became globally available to developers and enterprises through the API in AI Studio and the Gemini platform.[2] Google AI Ultra subscribers can access it in Antigravity; Google AI Ultra and Pro subscribers can use it in the Gemini app and AI Mode in Search.[2] The evidence therefore describes different access routes, rather than universal access across Google products. Before the expansion, Google said Argon was rolling out to a set of trusted cyber defenders through its Fairwind Program.[1]

Google set introductory API prices at $2 per million input tokens and $10 per million output tokens.[1] It priced cached input tokens at 95% below the input-token price.[1] Those figures give teams a starting point for estimating usage, though the evidence does not state what a typical project would cost. Google said it was participating in the U.S. government’s voluntary process for pre-release model access and would gather early-tester feedback and iterate on guardrails before expanding availability.[1] The company’s rollout plan therefore included both limited early access and later expansion.

Google says thousands of its employees highlighted Argon’s strengths in specialized coding, deeper research and writing quality.[1] The company also said Argon beat a published baseline by 40% in a matter of minutes, without specifying the baseline in the evidence.[1] For Argon in Antigravity, Google reported that teams increased agentic code submissions by more than 35% and reduced code rollbacks by nearly a third.[2] Google says Argon ranked first on the Vals Index, which covers sectors including finance, coding, legal and tax work, and helped a team refactor code across hundreds of files, completing months of work in weeks.[2]