Google has launched its new flagship artificial intelligence model, Gemini 4 Argon, making it available worldwide to developers and businesses from October 13, 2026. The release, announced by Google DeepMind senior vice president Koray Kavukcuoglu, marks the company's biggest AI moment of the year — and comes after months of delays, a leadership shakeup at DeepMind, and intensifying competition from OpenAI, Anthropic and Meta.
The story is trending per news coverage across international outlets, with Reuters, The Decoder and Google's own product blog all reporting on the launch. Here is a full explainer of what Argon is, what it costs, how to get it, and why this release matters.
What Google actually launched
Gemini 4 Argon is the top-tier model in Google's next AI generation. It is built for what Google calls "deep work" — complex, multi-step tasks across coding, finance, law, cybersecurity and science — where the model must plan, reason and execute across many steps rather than answer a single question. Google says its depth and precision across such tasks leads Argon to outperform other frontier models on benchmarks, ranking first on the Vals Index, a benchmark that measures model performance across critical economic sectors including finance, coding, legal and tax work.
Under the hood, Argon ships with a one-million-token context window (1,048,576 tokens shared between input and output), accepts text, images, video, audio and PDFs as input, and produces text output. Notably, it does not support image generation, audio generation or the Live API, according to Google's developer documentation — this is a text-reasoning specialist, not an all-media generator.
Pricing: cheap now, doubling on November 30
Argon launched with an introductory API price of $2 per million input tokens and $10 per million output tokens, with cached input tokens discounted 95 percent. But the discount has a deadline: after November 30, 2026, standard pricing will be $4 per million input tokens and $20 per million output tokens.
That positions Argon aggressively against rival frontier models on price — though developers planning long-term workloads should budget for the December price jump. Google's default synchronous output is capped at 65,536 tokens, with a continuation protocol that allows longer generations within the shared million-token budget.
Why this launch matters: delays and the DeepMind shakeup
The Argon launch lands after a rocky stretch for Google's AI program. The company announced in September that it no longer planned to release Gemini 3.5 Pro, which CEO Sundar Pichai had originally said would ship in June. Google then overhauled DeepMind after co-founder and CEO Demis Hassabis stepped aside, and several leaders of its Gemini model team left the company, as Reuters reported.
Against that backdrop, Argon is widely read as Google's bid to regain ground against rivals after months on the defensive. The model was first announced on September 30 with initial access limited to trusted cybersecurity partners; the October 13 rollout is the wide release that puts it in the hands of paying developers and enterprises everywhere.
Who is already using it
Google says early customers testing Argon include Deloitte, LexisNexis, Ambience Healthcare and Alphabet's own drug-discovery unit Isomorphic Labs. Teams using Argon inside Google's Antigravity coding tool increased agentic code submissions by more than 35 percent while reducing code rollbacks by nearly a third, Google claims. Its Chrome team reportedly used the model to restructure hundreds of code files in weeks rather than months, and cybersecurity firm Wiz reportedly used it to identify vulnerabilities affecting government and healthcare systems.
Separately, Google is opening its Fairwind Program to hundreds of additional partners. The program gives vetted cybersecurity defenders a version of Argon without its cyber guardrails — a striking move that treats the model's offensive-security potential as a capability to be shared with defenders. Google also said it would give 10,000 scientists six months of free access to Antigravity's highest usage tier.
How to access Gemini 4 Argon
Access is tiered, and the headline detail for everyday users is that Argon is not free. Developers and enterprises can use it through the Gemini API via AI Studio and the Gemini platform. Google AI Ultra subscribers get it in the Antigravity coding tool, while Google AI Ultra and Pro subscribers can access it through the Gemini app and AI Mode in Search. Free Gemini users and Plus subscribers, by contrast, now only have access to weaker Flash models under Google's new tier structure.
Google says Argon provides more direct answers, follows writing instructions better and avoids unnecessary repetition compared with previous Gemini models. On safety, Google claims Argon reduced successful prompt-injection attacks to 0.7 percent in internal testing.
The competitive landscape
Argon's wide release comes one day after Meta launched its own frontier model, Muse Wonder — covered earlier this week in Meta's Muse Wonder explained. Meanwhile, Anthropic continues to set the pace in independent evaluations: reporting by The Decoder notes that Argon still trails Anthropic's Claude Opus 5.5 in independent overall rankings, despite topping Google's preferred Vals Index. For more on Anthropic's recent AI-safety research, see Anthropic's warning on self-replicating AI, and for the latest on frontier-model capabilities elsewhere, OpenAI's new AI math results.
Google is not done: the company says more Gemini 4 models will follow — likely updated Flash variants that would eventually trickle down to free and Plus users. The Decoder also reports that Google is internally testing a potential successor codenamed "Carbon," said to outperform Argon at coding in particular. If that reporting holds, the flagship cycle is already accelerating.
What happens next
Three things matter from here. The introductory pricing ends November 30, 2026, when API rates double — a deadline that will shape early enterprise adoption. The expanded Fairwind rollout will put guardrail-free Argon in the hands of hundreds of cybersecurity partners, a move likely to draw scrutiny from AI-safety advocates. And with Meta, Anthropic and OpenAI all shipping frontier models within weeks of each other, the benchmark wars of late 2026 are just getting started.
Worldlession will update this story as more Gemini 4 models launch and pricing settles. The bottom line: Google has closed the gap in the frontier race, but the scoreboard — measured by independent benchmarks and developer adoption rather than launch-day claims — is still being written.