Google releases Gemini 4 Argon, its new flagship AI model
Google has released Gemini 4 Argon to subscribers, developers and businesses, claiming it matches or beats rival frontier models at lower prices. Coverage places the launch after months of delays and a leadership shake-up at DeepMind.
1 / 2
The story, neutrally told
Mixed · 2Google released its flagship AI model Gemini 4 Argon on 13 October 2026, saying it delivers "frontier reasoning and improved performance across complex workflows". The VergeLC “Google says Argon delivers “frontier reasoning and improved performance across complex workflows,”” Read at The Verge ↗ CNAN “released its flagship AI model Gemini 4 Argon to developers and businesses worldwide” Read at CNA ↗ Mixed · 2Google says Argon outperforms GPT-6 Astra, Claude Fable 5.1 and Claude Opus 5.5 on benchmarks including coding, cybersecurity, science, finance and law; these are the company's claims, and the articles do not report independent verification. The VergeLC “The company claims Argon outperforms rival models like GPT-6 Astra, Claude Fable 5.1, and Claude Opus 5.5 across a range of benchmarks” Read at The Verge ↗ CNAN “excels at "deep work" on complex, multi-step tasks in coding, finance, law, cybersecurity and science, Google said” Read at CNA ↗ Mixed · 2Google AI Pro and Ultra subscribers get Argon in the Gemini app and AI Mode in Search, while the Antigravity development tool is limited to Ultra subscribers. The Verge says it is unclear when, or if, Plus subscribers or free users will get it. The VergeLC “It is not clear when — or if — Argon will be available to Plus subscribers or free users.” Read at The Verge ↗ CNAN “Google AI Pro and Ultra subscribers can access it through the Gemini app and AI Mode in Search, while the Antigravity coding tool is limited to Ultra subscribers.” Read at CNA ↗
Mixed · 2Launch pricing is $2 per million input tokens and $10 per million output tokens, and the rates are set to double at the end of November (after 30 November, per CNA). The Verge puts the introductory price at around a fifth of the cost of OpenAI's GPT-6 Astra and Anthropic's Fable 5 models. It adds that per-token price does not necessarily mean lower real costs, because models use different numbers of tokens for the same task. The VergeLC “around a fifth the cost of OpenAI’s GPT-6 Astra and Anthropic’s Fable 5 models.” Read at The Verge ↗ CNAN “with rates set to double after November 30” Read at CNA ↗ Mixed · 2Google says it already uses Argon internally, to find and patch security vulnerabilities, to halve the computing power for YouTube's comments section, and to debug the drone physics simulator for Wing. It also says teams using Argon in Antigravity raised agentic code submissions by more than 35 per cent and cut code rollbacks by nearly a third. The VergeLC “halving the computing power needed to power YouTube’s comments section, and debugging the drone physics simulator for Alphabet’s drone delivery company Wing.” Read at The Verge ↗ CNAN “increased agentic code submissions by more than 35 per cent, while reducing code rollbacks by nearly a third, Google said.” Read at CNA ↗ Mixed · 2Before the public release, Argon went to a limited group of cybersecurity partners. Google now says it will open its Fairwind Program to hundreds more partners, giving vetted defenders a version of Argon without its cyber guardrails. Early testers include Deloitte, LexisNexis, Ambience Healthcare and Isomorphic Labs, and Google will give 10,000 scientists six months of free access to Antigravity's top usage tier. CNAN “The program gives vetted cybersecurity defenders a version of Argon without its cyber guardrails.”“Customers testing the model early include Deloitte, LexisNexis, Ambience Healthcare and Isomorphic Labs.” Read at CNA ↗ The VergeLC “the public rollout follows a limited release to a select set of “trusted cyber defenders”” Read at The Verge ↗
Mixed · 2The launch follows a difficult period for Google's AI work. Google announced in September that it no longer planned to release Gemini 3.5 Pro, which CEO Sundar Pichai had said would arrive in June. DeepMind was overhauled after co-founder and CEO Demis Hassabis stepped aside and several Gemini leaders left. The Verge calls Argon Google's first major frontier release since Gemini 3 in late 2025, and says it landed a day after Meta announced its own frontier model, Muse Wonder Full. CNAN “The company announced in September that it no longer planned to release Gemini 3.5 Pro”“Demis Hassabis stepped aside and several leaders of its Gemini model left the company” Read at CNA ↗ The VergeLC “Argon is Google’s first major frontier release since Gemini 3 in late 2025.” Read at The Verge ↗
Every sentence links to the reporting it rests on. The pill in front of each says where its sources sit: Left, Centre or Right when one side supplies at least half of them, Mixed when they are evenly split. The number is how many outlets it cites.
Left1 outlet
- Framing
- The Verge treats the launch as a comeback story: Google was trailing rivals and Argon shows it is still in the race.
- Emphasis
- Google's competitive position, benchmark claims against named rivals, the price advantage, and the caveat that per-token price may not mean lower real cost.
- Leaves out or plays down
- Does not mention the Gemini 3.5 Pro cancellation or the DeepMind leadership change by name, and gives no customer names.
- Charged language
- “finally”“Seismic internal shakeups”“far from a two-horse race”
Centre1 outlet
- Framing
- CNA runs a business-news account of a product launch aimed at developers and businesses, set against Google's recent setbacks.
- Emphasis
- Enterprise use, API pricing, early customers, Google's productivity figures, the Fairwind cyber programme, and the Gemini 3.5 Pro cancellation and DeepMind overhaul.
- Leaves out or plays down
- Does not name the rival models Google compares itself against, and does not mention the competing Meta release or the caveats on per-token pricing.
- Charged language
- “rocky stretch”
Right0 outlets
No right outlet in our sources has covered this story yet.
What every side reports
- Google released Gemini 4 Argon on 13 October 2026.
- Launch API pricing is $2 per million input tokens and $10 per million output tokens, set to double at the end of November.
- Pro and Ultra subscribers get Argon in the Gemini app and AI Mode in Search, and Antigravity access is limited to Ultra subscribers.
- Argon was first given to select cybersecurity partners before the public release.
Where accounts differ
-
Release date
- Left
- The Verge says it is rolling out "today" (13 October).
- Centre
- CNA's text says Google released the model "on Wednesday", though its dateline is Oct 13.
Google organisation
Google presents Argon as a frontier model that beats rivals on benchmarks, backs this with internal use cases and early customers, and prices it competitively. It is widening access to a cyber-unrestricted version for vetted defenders.
“Google says it’s already putting Argon to use internally” — The Verge
“Google will offer 10,000 scientists six months of free access to Antigravity's highest usage tier.” — CNA
Gemini 4 Argon product
Gemini 4 Argon is described as Google's flagship model, aimed at complex multi-step work in coding, finance, law, cybersecurity and science. It is available first to paid tiers.
“The model excels at "deep work" on complex, multi-step tasks” — CNA
“The model is rolling out today to Google’s higher-tier Pro and Ultra subscribers” — The Verge
Left1 article
-
Google finally releases Gemini 4 Argon
Mixed Analytical take that presents Google's claims as claims, frames the launch as a comeback, and adds caveats on pricing.
Centre1 article
-
CNAN ·
Google launches flagship AI model Gemini 4 Argon for developers, businesses
Neutral Business report on the launch that attributes performance figures to Google and sets out the company's recent troubles.
Right0 articles
No coverage yet.
- 13 Oct 17:00 First The VergeLC Google finally releases Gemini 4 Argon
- 13 Oct 17:01 +2m CNAN Google launches flagship AI model Gemini 4 Argon for developers, businesses
Times are when each article was published, or when we first saw it if the outlet gave no time.
Ask about this story
Loading…