Google launches Gemini 4 Argon flagship AI model
Google has launched Gemini 4 Argon across its API and subscription tiers, touting top benchmark results in knowledge work and coding despite mixed refactoring performance in early evaluations.
Google released Gemini 4 Argon globally, representing its first major flagship model upgrade since Gemini 3 Pro last November. The new model is available through the paid tier of the Gemini API at $2 per million input tokens and $10 per million output tokens, supporting an input limit of 1,048,576 tokens and an output limit of 32,768 tokens with thinking enabled. For consumer applications, availability is restricted to paying subscribers, rolling out to AI Pro and AI Ultra tiers in the Gemini app and Search, and to Ultra subscribers inside Google's Antigravity desktop agent app.[1][2][3][4][6][7][10][11][12][14]
Google is positioning Argon as a workhorse model tailored for complex tasks across coding, cybersecurity, finance, law, and science. A benchmark comparison shared alongside the release shows Argon leading rival models—including GPT-6 Astra, Claude Fable 5.1, and Claude Opus 5.5—on benchmarks such as the Vals Index (68.9%), DeepSWE v1.1 (77.9%), LABBench 2 (88.8%), and LVBench (91.7%). However, GPT-6 Astra maintained the lead on FrontierSWE v2 (65.5% versus Argon's 55.0%), Terminal-Bench Science 0.1 (68.1% to 57.6%), and the OSWorld-2.0 offline subset (72.6% to 69.2%).[5][8][9][13]
Early third-party testing highlighted sharp contrasts in model performance. According to Dan Shipper and a team of testers at Every, Gemini 4 Argon proved to be an exceptional writing partner—benefiting from DeepMind hiring writers to curb AI slop—and successfully built a working webcam heart-rate monitor in 35 minutes. Conversely, the model struggled significantly with existing codebases. On Shipper's Senior Engineer refactoring benchmark, Argon scored just 21% over four hours and 47 minutes, compared to 81% to 85% for competing frontier models, while also skipping inspection steps during code reviews in 17 of 38 benchmark runs.[12]
Key facts
- Gemini 4 Argon is Google's first major flagship model release since Gemini 3 Pro launched in November 2025.
- The model is priced at $2 per million input tokens and $10 per million output tokens in the Gemini API, with support for over 1 million input tokens.
- End-user access is restricted to paid subscriptions, rolling out to AI Pro and AI Ultra tiers in Search and the Gemini app, and Ultra users in Antigravity.
- Rollout benchmark figures show Argon leading rivals on Vals Index (68.9%) and DeepSWE v1.1 (77.9%), but trailing GPT-6 Astra on FrontierSWE v2 (55.0% vs. 65.5%) and OSWorld-2.0 (69.2% vs. 72.6%).
- On Every's Senior Engineer refactoring benchmark, Gemini 4 scored 21%, trailing GPT-6 Astra, Claude Fable 5.1, and Claude Opus 5.5, which scored between 81% and 85%.
- Google positioned Gemini 4 Argon to handle complex professional tasks across coding, cybersecurity, finance, law, and science.
Sources · 13 sources
- �A
🚨 AI News | TestingCatalog@testingcatalogPost on X ·
BREAKING 🔥: Google has officially announced Gemini 4 Argon to be available on the APIs and on Gemini for Ultra subscribers. > "For developers and enterprises, it’s available in the paid tier of the Gemini API and the Gemini platform, for Ultra subscribers in Google Antigravity, and Ultra and Pro subscribers in the Gemini app and Search." Ultra testing time! 👀 https://t.co/UsGK0qeqDP
Open source - 9T
9to5Mac@9to5macPost on X ·
Gemini app rolling out Gemini 4 Argon to AI Pro and AI Ultra https://t.co/EvM9CbnIwU by @technacity
Open source - TE
Techmeme@TechmemePost on X ·
Google rolls out Gemini 4 Argon globally via the Gemini API and to Google AI Pro subscribers in the Gemini app and AI Mode, and in Antigravity for Ultra users (@theroberthart / The Verge) (Visit Techmeme dot com for the link and full context!)
Open source - AT
Ars Technica@arstechnicaPost on X ·
Gemini 4 Argon begins rolling out in Gemini app and Antigravity, but only if you pay https://t.co/XFkLX7JGmS
Open source - CB
Crypto Briefing@Crypto_BriefingPost on X ·
🤖NEW: Google launches Gemini 4 Argon, its new frontier AI model for complex work across coding, finance, law, cybersecurity and science. https://t.co/bPJaKZYXAd
Open source - LA
Lisan al Gaib@scaling01Post on X ·
Gemini 4 Argon is in the Gemini API https://t.co/Bp4ctY95G8
Open source - CB
Crypto Briefing@Crypto_BriefingPost on X ·
🤖NEW: Google launches Gemini 4 Argon, its new frontier AI model for complex work across coding, finance, law, cybersecurity and science. Argon supports up to 1M tokens and starts at $2 per million input tokens and $10 per million output tokens. https://t.co/jH0YnJYc2J
Open source - CB
Crypto BriefingArticle ·
Google launches new frontier model Gemini 4 Argon with deeper reasoning Google's Gemini 4 Argon model enhances AI's role in complex professional tasks, potentially transforming industries like finance and science. The post Google launches new frontier model Gemini 4 Argon with deeper reasoning appeared first on Crypto Briefing .
Open source - BI
Business Insider@BusinessInsiderPost on X ·
Google is positioning Argon as its new workhorse model, stating that it excels at coding, cybersecurity, and knowledge work in finance and law. https://t.co/qYGrs5g5XG
Open source - DH
Demis Hassabis@demishassabisPost on X ·
Gemini 4 Argon is amazing for knowledge work and coding - and it’s super fun to build with! We asked it to create this wizard chess game and it's so cool to see Argon's physics understanding & 3D modelling capabilities, makes me nostalgic for Battlechess on the Amiga - fun times! Argon is rolling out today to Google Ultra and Pro subscribers in the @GeminiApp and Search, and Ultra subscribers in @antigravity! More info & full availability in the blog: https://t.co/7d3791mRNH
Open source - EV
Evan@StockMKTNewzPost on X ·
Google $GOOGL CEO Sundar Pichai just posted this: “Gemini 4 Argon is here.” https://t.co/S53BVw0CHF
Open source - DS
Dan Shipper@danshipperPost on X ·
BREAKING: Gemini 4 is OUT and it's SURPRISINGLY GOOD?? I admit I was ready to trash this model qhen I saw the benchmarks but six of us at @every spent the past week testing it across coding, writing, design, and knowledge work, and were pleasantly surprsied. It’s a breakthrough if your company lives in Google’s ecosystem, and a strong option for frontier hoppers as a secondary model for certain tasks. I Here’s our vibe check: My new favorite writing partner. DeepMind hired professional writers with red pens to stamp out AI slop, and it worked. I use it to untangle structure and sharpen metaphors; it also taught me something new tracing in a few of my favorite books. It's got a natural feel for structure and detailed explanations that Astra and Opus lack. Shaky at refactors, fast on greenfield apps. At low effort on a from-scratch build, Gemini 4 built @kieranklaassen a working webcam heart-rate monitor in 35 minutes—matching Claude Opus 5.5’s quality and beating OpenAI’s models on that test. Where it cuts corners is self-verification and existing codebases. In 17 of Kieran’s 38 benchmark runs, it opened and closed its code-review step just 30 milliseconds apart instead of inspecting its changes. And on my Senior Engineer benchmark—refactoring a vibe coded slop codebase—it worked for four hours and 47 minutes and scored just 21% (versus 81% to 85% for GPT-6 Astra, Fable 5.1, Sonnet 5.5, and Opus 5.5). Not great. A 2026 model trapped in a 2025 harness. Antigravity, Google’s desktop agent app, feels six months to a year behind Claude Code and Codex—lacking cloud environments, SSH, a built-in browser, and smart permission checks (leaving you clicking constant pop-ups or using unrestricted "Turbo" mode). Net Result: I keep ChatGPT and Claude open day to day, and switch to Gemini 4 when I need an editorial partner to structure a tough essay, sharpen a metaphor, or close-read a book. Reach for it for live writing collaboration, fast low-effort prototypes, or steerable UI and dashboard design—ideally via Cursor or the API to skip Antigravity. When a model is good enough to force you to use a bad harness, it tells you something. State of Play: Google hadn’t shipped a major flagship upgrade since Gemini 3 Pro last November. For teams whose companies live in Google’s ecosystem, the model you can use inside your approved work tools finally belongs with the frontier pack. DeepMind closed the model gap; now Google needs to bring the app around it into 2026. Read our full vibe check: https://t.co/ZRITq2iFyl
Open source - MT
MTS@MTSlivePost on X ·
SITUATION DETECTED: Google's Gemini 4 Argon is rolling out now. https://t.co/MAr3iaZfqf
Open source - TN
The New Stack@thenewstackPost on X ·
Two weeks after a cyber-defenders-only debut, Google’s new flagship hits the Gemini API at $2/$10 per million input/output tokens. https://t.co/Y5ypU77lMF
Open source