conv.

All stories
AIActive today · day 8

Google launches Gemini 4 Argon, its first frontier model since Gemini 3

Google rolls out Gemini 4 Argon globally behind a paywall, claiming parity with GPT-6 Astra on coding, robotics and cyber-defense benchmarks.

newest coverage 9h agosummary written 17h ago228 pieces4.6k likes & upvotes1.2k comments Google NewsRedditHacker NewsMastodonNewswiresXYouTube

Part of a larger narrative

Google Gemini Security Vulnerabilities

3 stories · since Sep 23 · newest 9h ago — Google's Gemini AI system has demonstrated security risks through hacking incidents during testing while the company simultaneously advances its capabilities…

  1. Google launches Gemini 4 Argon, its first frontier model since Gemini 3you are here
  2. Google shuts down Gemini Gems, shifts to 'skills' amid AI agent competition
  3. Google launches Gemini 4 Argon, locks it behind 'trusted cyber defenders' access

What to know

  • Gemini 4 Argon is Google's first frontier-model release since Gemini 3 in late 2025, arriving after months of the company trailing Anthropic and OpenAI, and one day after Meta's Muse Wonder Full.
  • Independent benchmarking (Artificial Analysis) found Argon ties OpenAI's GPT-6 Astra on intelligence while costing about 60% as much per task during an introductory discount through November 30.
  • Access is gated: available to Pro/Ultra subscribers in the Gemini app and AI Mode, and only to Ultra users in the Antigravity dev platform, drawing complaints about Google's confusing rollout and paywall.
  • Developer testing is mixed: some report a clear step up in writing and agentic coding performance, while others rank Argon behind Claude Opus 5.5 and GPT-6.1 Sol, roughly two months behind the frontier.

The dispute Whether Argon's benchmark claims of frontier parity hold up in practice, versus hands-on testers finding it a tier below Opus 5.5 and GPT-6.1 Sol. · positions read across 115 posts and comments

most voices

Argon represents a genuine step up in writing quality, reasoning, and general usefulness.

  • “This is a clear step change, I'm genuinely amazed. Its answers are to-the-point, logically structured, clearly written. Like a really well made mini essay on your question.”

    svara · Hacker News ↗
many voices

Argon is solid but still trails the true frontier (Opus 5.5, GPT-6.1 Sol/Astra) in coding benchmarks.

  • “It's roughly in between Fable 5 and Fable 5.1 in our testing, so it's maybe 2 months behind the frontier.”

    gertlabs · Hacker News ↗
some voices

Google's rollout and access gating are poorly executed and confusing.

  • “Classic Google: Clicks the "Try in Gemini" button from the Gemini 4 page, gets a page that defaults to Gemini 3.6, with no option to try Gemini 4 without upgrading.”

    vidarh · Hacker News ↗

“frontier reasoning and improved performance across complex workflows”

Google, Company statement · The Verge · Oct 12

Google / Google DeepMind Developer of Gemini 4 ArgonLogan Kilpatrick Google (OfficialLoganK)Artificial Analysis Independent AI benchmarking firmOpenAI Rival AI lab (GPT-6 Astra)Anthropic Rival AI lab (Claude Fable/Opus)

Google launches Gemini 4 Argon, its first frontier model since Gemini 3
x.com

How it unfolded 7 developments, newest first · click a bar or a number to jump articlesvideospostscomments

Peak 145 pieces in two hours at Oct 13, 11 AM; 228 pieces over 8 days (89 articles · 1 video · 55 posts · 83 comments) Oct 7, 9 AM — 1 piece · 1 article — Google News 1Oct 7, 11 AM — quietOct 7, 1 PM — quietOct 7, 3 PM — quietOct 7, 5 PM — quietOct 7, 7 PM — quietOct 7, 9 PM — quietOct 7, 11 PM — quietOct 8, 1 AM — quietOct 8, 3 AM — quietOct 8, 5 AM — quietOct 8, 7 AM — quietOct 8, 9 AM — quietOct 8, 11 AM — quietOct 8, 1 PM — quietOct 8, 3 PM — quietOct 8, 5 PM — quietOct 8, 7 PM — quietOct 8, 9 PM — quietOct 8, 11 PM — quietOct 9, 1 AM — quietOct 9, 3 AM — quietOct 9, 5 AM — quietOct 9, 7 AM — quietOct 9, 9 AM — quietOct 9, 11 AM — quietOct 9, 1 PM — quietOct 9, 3 PM — quietOct 9, 5 PM — quietOct 9, 7 PM — quietOct 9, 9 PM — quietOct 9, 11 PM — quietOct 10, 1 AM — quietOct 10, 3 AM — quietOct 10, 5 AM — quietOct 10, 7 AM — quietOct 10, 9 AM — quietOct 10, 11 AM — quietOct 10, 1 PM — quietOct 10, 3 PM — quietOct 10, 5 PM — quietOct 10, 7 PM — quietOct 10, 9 PM — quietOct 10, 11 PM — quietOct 11, 1 AM — quietOct 11, 3 AM — quietOct 11, 5 AM — quietOct 11, 7 AM — quietOct 11, 9 AM — quietOct 11, 11 AM — quietOct 11, 1 PM — quietOct 11, 3 PM — quietOct 11, 5 PM — quietOct 11, 7 PM — quietOct 11, 9 PM — quietOct 11, 11 PM — quietOct 12, 1 AM — quietOct 12, 3 AM — quietOct 12, 5 AM — quietOct 12, 7 AM — quietOct 12, 9 AM — quietOct 12, 11 AM — quietOct 12, 1 PM — quietOct 12, 3 PM — quietOct 12, 5 PM — quietOct 12, 7 PM — quietOct 12, 9 PM — quietOct 12, 11 PM — quietOct 13, 1 AM — quietOct 13, 3 AM — 2 pieces · 1 post · 1 comment — Reddit 2Oct 13, 5 AM — quietOct 13, 7 AM — quietOct 13, 9 AM — quietOct 13, 11 AM — 145 pieces · 80 articles · 1 video · 43 posts · 21 comments — Newswires 70, Hacker News 28, X 26, +4 moreOct 13, 1 PM — 24 pieces · 4 articles · 6 posts · 14 comments — Hacker News 17, Newswires 4, X 2, +1 moreOct 13, 3 PM — 17 pieces · 2 articles · 1 post · 14 comments — Hacker News 14, Google News 1, Newswires 1, +1 moreOct 13, 5 PM — 8 pieces · 1 post · 7 comments — Hacker News 8Oct 13, 7 PM — 3 pieces · 1 post · 2 comments — Hacker News 2, Mastodon 1Oct 13, 9 PM — 2 pieces · 2 comments — Hacker News 2Oct 13, 11 PM — 5 pieces · 2 articles · 3 comments — Hacker News 3, Google News 1, Newswires 1Yesterday, 1 AM — 2 pieces · 1 post · 1 comment — Hacker News 1, X 1Yesterday, 3 AM — 8 pieces · 8 comments — Hacker News 8Yesterday, 5 AM — 2 pieces · 1 post · 1 comment — Hacker News 1, Mastodon 1Yesterday, 7 AM — 3 pieces · 3 comments — Hacker News 3Yesterday, 9 AM — 3 pieces · 3 comments — Hacker News 3Yesterday, 11 AM — 1 piece · 1 comment — Hacker News 1Yesterday, 1 PM — 1 piece · 1 comment — Hacker News 1Yesterday, 3 PM — 1 piece · 1 comment — Hacker News 1Yesterday, 5 PM — quietYesterday, 7 PM — quietYesterday, 9 PM — quietYesterday, 11 PM — quietToday, 1 AM — quiet ◂ 1 earlier2–67
Oct 8Oct 9Oct 10Oct 11Oct 12Oct 13yesterdaynow · 2:52 AM ET
  1. 7

    Argon added to OpenCode Go with discounted usage

    Coding assistant OpenCode announced Argon's availability in OpenCode Go, offering $60 of usage for $10 until November 30.

    “Gemini 4 Argon is now available in OpenCode Go. $60 usage for $10 til Nov 30…”
    — @opencode

    • Gemini 4 Argon is now available in OpenCode Go $60 usage for $10 til Nov 30

      @opencodeX23h ago928▲view on X ↗
    2 more of the top 3 · 26 posts in this stretch
    • Arint@arint.info

      RT @opencode: Gemini 4 Argon ist jetzt in OpenCode Go verfügbar mehr auf Arint.info # AI # CloudComputing # Gemini4Argon # Nutzung # OpenCodeGo # TechNews # arint_info

      Arint@arint.infoMastodon20h ago1▲view on Mastodon ↗

    • We evaluated Gemini 4 Argon in 100 unsaturated, competitive and cooperative coding environments. It's roughly in between Fable 5 and Fable 5.1 in our testing, so it's maybe 2 months behind the frontier.Google's major releases have usually done pretty well on our benchmarks because they're good at hard problems and open-ended decision making (with…

      gertlabsHacker News23h agoview on Hacker News ↗
    all of them →
  2. 6

    Access confirmed limited to paying subscribers

    Reporting clarified Argon is gated behind paid tiers: Pro/Ultra in the Gemini app and Ultra-only in Antigravity, with developers getting a 50% discounted API rate through November.

    “Classic Google: Clicks the "Try in Gemini" button from the Gemini 4 page, gets a page that defaults to Gemini 3.6, with no option to try Gemini 4 without upgrading.”
    — vidarh, Hacker News commenter · source
    1. 2 outlets Gemini 4 Argon is finally available for developers, businesses, and paid users

      first by Digital Trends, 1d ago · also Google

      1 more headline
    2. 1 outlet Gemini 4 Argon is finally here, and it’s not just for coders and developers

      first by Android Authority, 1d ago

    • Gemini 4 Argon is now available for PRO subs, but only though the Gemini App. Trying to build something cool with it.

      @MiaAI_labX1d ago9▲view on X ↗
    2 more of the top 3 · 26 posts in this stretch

    • It’s wild how good it is at explaining nuanced and complex health topics.As an example, I recently had an OCT scan of my retina because I'm a “glaucoma suspect.” I'm a glaucoma suspect because I have an unusually high cup-to-disc ratio, which is a pattern typically seen in glaucomatous patients.As part of this OCT scan, you get a report back on…

      sebmellenHacker News1d agoview on Hacker News ↗

    • Classic Google: Clicks the "Try in Gemini" button from the Gemini 4 page, gets a page that defaults to Gemini 3.6, with no option to try Gemini 4 without upgrading.If they don't want to make it available even for small tests for free, that's fine, but then don't put a button suggesting I can try it without qualifying it...EDIT: It's actually…

      vidarhHacker News1d agoview on Hacker News ↗
    all of them →
  3. 5

    Users discover Argon's January 2025 knowledge cutoff

    Reddit users flagged that despite its October 2026 release, Argon's training data knowledge cutoff is January 2025.

    • I'm a robotics noob currently building a prototype of a snow sweeping robot (www.frost-e.com) using my own design and I'm encouraged to see similar work coming from real places. Namely:- Zero shot control using a visual harness and action primitives. I've seen this from a few other places as well (https://tml.stanford.edu/homebody/) and it's…

      martythemaniakHacker News1d agoview on Hacker News ↗
    2 more of the top 3 · 7 posts in this stretch

    • How many cicles before pelicans become photoreallistic SVGs and them come back to stylistic artistic abstract renderings?Also, that first tweet contains a commentary that, for me, explain a lot about the public misconception about how current AI works:> wait, does this mean you guys are actively training svg generation? i was mostly amazed by it…

      motoboiHacker News1d agoview on Hacker News ↗

    • I am very new to robots and recently decided to go crazy and buy a humanoid from Unitree. I've spent a while building some tools in order to integrate LLMs with vision capabilities that can control the robot motion in a simple manner... and while I saw Opus and Astra crashing against tables, beds and surfaces (yeah, my LIDAR data isn't being used…

      comandillosHacker News1d agoview on Hacker News ↗
    all of them →
  4. 4

    Independent benchmark ties Argon with GPT-6 Astra at lower cost

    Artificial Analysis found Argon matches GPT-6 Astra on its Intelligence Index while costing about 60% as much per task during the introductory discount period, marking Google's return to the frontier.

    “Gemini 4 Argon is now GA. Argon ranks equal to GPT-6 Astra on the Artificial Analysis Intelligence Index at 60% of the Cost per Task with discounted prices until Nov 30…”
    — @ArtificialAnlys

    • Gemini 4 Argon is now GA. Argon ranks equal to GPT-6 Astra on the Artificial Analysis Intelligence Index at 60% of the Cost per Task with discounted prices until Nov 30 Key takeaways: - Google is back to having a frontier model released to the public: This is Google's first

      @ArtificialAnlysX1d ago1.3k▲view on X ↗
    2 more of the top 3 · 10 posts in this stretch

    • As a cyclist, I tend to pay more attention to the bike shape than most people. Everyone knows what a bike looks like, but as with LLMs, most people would struggle to draw one - correctly connecting all the different tubes at correct angles etc.I note that even these bikes are not actually rideable. For a bike to be stable, the front wheel axle…

      rich_sashaHacker News1d agoview on Hacker News ↗
    • slashdot@mastodon.cloud

      Google Launches Flagship AI Model Gemini 4 Argon https:// tech.slashdot.org/story/26/10/ 13/1643231/google-launches-flagship-ai-model-gemini-4-argon?utm_source=rss1.0mainlinkanon

      slashdot@mastodon.cloudMastodon1d agoview on Mastodon ↗
    all of them →
  5. 3

    Google details Argon's robotics and physical-reasoning capabilities

    A companion blog post showed Argon scoring 87% on the Embodied Reasoning (ERQA) benchmark and being used to control robotic arms, quadrupeds and humanoids via digital twins and closed-loop visual reasoning.

    “I'd put it at B tier, just under gpt 6.1 sol. It isn't A tier like Opus 5.5 or Astra.”
    — jjcm, Hacker News commenter · source

    • Image->html test for Argon. I'd put it at B tier, just under gpt 6.1 sol. It isn't A tier like Opus 5.5 or Astra.Argon build: https://html.non.io/lcars-argon/GPT 6.1 sol: https://html.non.io/lcars-gpt-6.1-sol/Opus 5.5 (currently SOTA for this): https://html.non.io/lcars-opus-5.5/Original designs:

      jjcmHacker News1d agoview on Hacker News ↗
    2 more of the top 3 · 11 posts in this stretch
    • newsyc200@mastodon.social

      Gemini 4 Argon: https:// deepmind.google/models/gemini/ Discussion: http:// news.ycombinator.com/item?id=5 0073879

      newsyc200@mastodon.socialMastodon1d agoview on Mastodon ↗

    • simonw, your pelicans are so famous now that the frontier labs can just RLVR for SVG images explicitly.It's a simple manner of prompt...generate concept art with nanobanana, measure, llm creates svg with text, measure the distance between rasterized svg embedding to nanobanana art....do post-training RL on a 3 million such images and matching…

      LarsDu88Hacker News1d agoview on Hacker News ↗
    all of them →
  6. 2

    Google officially launches Gemini 4 Argon globally

    Google rolled out Gemini 4 Argon via the Gemini API, the Gemini app and AI Mode in Search for Pro/Ultra subscribers, and in Antigravity for Ultra users, positioning it as a frontier model for coding, math, science and cyber defense.

    “Today, Gemini 4 Argon is officially rolling out.”
    — @Google

    1. 12 outlets first by Tech Monitor, 1d ago · also Bitcoin Insider, ANI News, CoinGape, Chrome Unboxed, Financial Express, NPowerUser +5 · read ↗

    2. 10 outlets Google's Gemini Argon 4 Is Now Available Worldwide

      first by Gizmodo, 1d ago · also Deal N Tech, Neowin, Google Antigravity, Devin Blog, CNET, Business Insider +3

      8 more headlines
    9 more claims →

    • Today, Gemini 4 Argon is officially rolling out.

      @GoogleX1d ago648▲view on X ↗
    2 more of the top 3 · 35 posts in this stretch

    • Unsurprisingly, the pelicans are excellent. Google are the only lab who have publicly shared that they have specifically trained for SVG output - https://twitter.com/sunjiao123sun_/status/202455551655137292... - and they've had great pelican results in the past too (which they have shown off themselves -…

      simonwHacker News1d agoview on Hacker News ↗
    • rustybrick@c.im

      Gemini 4 Argon now powers AI Mode in Google Search https://www. seroundtable.com/google-ai-mod e-gemini-4-argon-42289.html

      rustybrick@c.imMastodon1d agoview on Mastodon ↗
    all of them →
  7. 5 days quiet
  8. 1

    Early reports predict imminent Argon launch

    Ahead of the official release, coverage speculated Gemini 4 Argon could launch that day, circulating anticipated specs and benchmarks.

    1. 1 outlet Gemini 4 Argon Could Launch Today: Full Specs, Benchmarks & Features

      first by Nokiapoweruser, 7d ago

Also covered reported alongside — the timeline has no entry for these yet

  1. 4 outlets Google Opens Gemini 4 Argon to Paying Users at $2 Per Million Input Tokens

    first by Implicator.ai, 1d ago · also Windows Report, VentureBeat, The New Stack

    3 more headlines
  2. 4 outlets Google Gemini 4 Argon rollout begins: Price, features and how to get the AI that's ‘built to solve your hardest problems’

    first by The Economic Times, 1d ago · also Google, Google DeepMind, Google AI Blog

    1 more headline

and 2 smaller pieces

What people are saying 10 voices from 3 sites · best of 115 · verbatim

Still unanswered
  • Will Argon ever reach free or Plus-tier users, or remain restricted to Pro/Ultra subscribers?
  • Does the January 2025 knowledge cutoff meaningfully limit real-world usefulness compared to rivals?