Gemini 4 Argon für zahlende Nutzer freigeschaltet: Was kann das neue Modell?
Google schaltet das nächste Flaggschiff-Modell frei: Analyse der neuen Fähigkeiten und Kosten. --- Im Schweizer Hochlohnland entscheidet die Effizienz von A…
Google schaltet das nächste Flaggschiff-Modell frei: Analyse der neuen Fähigkeiten und Kosten.
Warum dieser Ansatz für die Schweiz und den DACH-Raum entscheidend ist
Im Schweizer Hochlohnland entscheidet die Effizienz von Arbeitsabläufen über Wettbewerbsfähigkeit. Gleichzeitig verlangt das revidierte Schweizer Datenschutzgesetz (revDSG) strikte Kontrolle über vertrauliche Daten.
Statt für dutzende einzelne SaaS-Abos hunderte Franken pro Monat auszugeben, setzen führende Entwickler und smarte Betriebe 2026 auf Open-Source-Router, lokale Modelle und standardisierte Schnittstellen:
- 0 Fr. laufende Kosten: Maximale Nutzung der grosszügigen Free-Tiers und Open-Weights.
- 100% Datensouveränität: Sensible Keys und Abfragen bleiben verschlüsselt auf deiner eigenen Hardware.
- Kein Vendor-Lock-in: Fällt ein Anbieter aus, wechselt der Router in Millisekunden nahtlos zum nächsten Modell.
Schritt-für-Schritt Anleitung
Gemini 4 Argon is now open to paying users
(but your free account isn't invited)
In partnership with
Before we start… Google got it right with Argon. Gmail is another story, at least when it comes to your data. If you want an inbox only you can read, here's a free one:
Privacy-first email. Built for real protection.
Proton Mail offers what others won’t:
End-to-end encryption by defaultZero access to your dataOpen-source and independently auditedBased in Switzerland with strong privacy lawsFree to start, no ads
End-to-end encryption by default
Zero access to your data
Open-source and independently audited
Based in Switzerland with strong privacy laws
Free to start, no ads
We don’t scan your emails. We don’t sell your data. And we don’t make you dig through settings to find basic security. Proton is built for people who want control, not compromise.
Simple, secure, and free.
Explore Proton’s benefits
Your 30 second cheat sheet
What happened: Google's top model, Gemini 4 Argon, started rolling out yesterday in the Gemini app, AI Mode in Search and the API (Google's announcement).
Price: Google AI Pro at $19.99 a month, or Ultra from $99.99 (plans). Free accounts don't get it.
Use it when: a long document or a coding job keeps tripping up your current AI.
Skip it when: you just need quick answers. Free tools still handle those fine.
The model you were told to wait for just showed up
Two weeks ago, Google's strongest model was here but locked. Only vetted security teams could use it.
That changed yesterday. Google is now rolling Argon out to AI Pro and Ultra subscribers in the Gemini app and in AI Mode in Search. Developers on the paid API tier get it too.
So what's different for you? Google says Argon is better at long, multi-step jobs, the kind where other models lose the thread halfway.
It also writes more directly, with less hedging, and sticks closer to the style you ask for.
Fewer "great question!" openers. Fewer paragraphs that say nothing.
This is what it has achieved
Google shared four results from its own teams. They're Google's reports, but for most of them the code is public, so you can check.
A big Python project, sped up overnight. Given a one paragraph goal, Argon worked alone overnight on Google's open source Agent Development Kit, a toolkit for building agents. Its fixes cut session prep time by 40% in long sessions.
A chemistry tool, up to 2.3x faster. Scientists use xTB to predict a molecule's 3D shape. Argon's speedups are now in its GitHub repo.
Gemma, nearly 3x faster. In a few days, Argon took Google's small open model from about 37,900 to over 108,000 tokens a second, with no drop in quality. Google says that usually takes months. The code changes are public.
3D scenes and games in a browser tab. Googlers use it to prototype 3D visuals and animations in plain code.
— # (#)
— # (#)
Quick break from Gemini.Google’s engineers went to bed and woke up to faster code. Here’s how to get the same treatment for your reports and leads:
Your agents work while you sleep
Give a Skydive agent an ongoing responsibility and they’ll handle it on schedule, every time.
Have them prep your morning report, research new leads, monitor customer feedback, or keep projects moving overnight. You wake up, the work is already done.
Make your first agent
Hand it the job your AI keeps fumbling
The quickest way to know if Argon is worth $19.99 a month: give it a task your current AI got wrong.
Say ChatGPT botched the summary of a 30 page client contract last month. Paste the same contract into Argon with the same prompt and see which one catches the clauses that matter.
Step 1: Open Gemini and check the model picker. The rollout is gradual, so it might take a day to appear.
Step 2: Run the same prompt and files through both models. Decide what a good answer looks like before you start, so you judge results and not vibes.
Step 3: If you build with the API, the launch price is $2 per million input tokens and $10 per million output (a token is about three quarters of a word). It doubles after November 30, so test now. Setup is in the developer guide.
Something to watch out: AI Pro doesn't get Argon in AI Studio or Antigravity, Google's coding tools.
Those need Ultra, per Ars Technica's report. And Google's plans page also lists Argon under the $4.99 AI Plus plan, while the announcement only names Pro and Ultra.
Check your picker before upgrading.
Also happened yesterday
Facebook now takes requests. Meta's new Your algorithm tool lets you type the topics you want in Feed and Reels, and AI reads what you mean. If you post there, make your topic obvious early, because viewers can now ask for more or less of it in plain words.
A strong open model that runs on a laptop. Liquid AI released LFM2.5-24B-A2B, built for agents and tool use. Liquid says it hits about 184 tokens a second on an Apple M5 Max, and you can load it in LM Studio.
Anthropic outgrew Git. Merged code changes at the company went from about 2,500 a day in May to over 30,000 by September, so it built its own replacement, called Scion. You can't install it. It just shows how fast agent-written code is piling up.
You’re only {{rp_num_referrals_until_next_milestone}} referrals away from {{rp_next_milestone_name}}. Send this to that friend who pays for Gemini:
{{rp_refer_url}}
See you tomorrow.
Stefano MartellCo-Founder, The AI Leverage by Bestapps.ai
PS: Google just had a big week, but the lead seems to change hands every month. Pick yours:
Who's winning the AI race right now?
OpenAIGoogleAnthropicMetaxAIMistralDeepSeek
Login or Subscribe to participate
Keep Reading
The AI Leverage Newsletter
Praxistipps für Schweizer KMU & Power-User
- Verschlüsselte Schlüsselverwaltung: Speichere keine Produktiv-Keys in ungesicherten Umgebungsvariablen. Nutze lokale Secrets-Manager.
- Fallback-Hierarchie: Konfiguriere schnelle Modelle (z.B. Groq / Llama 3) für Routine-Klassifizierungen und schwere Modelle nur für komplexe Synthesen.
- DSG-Konformität: Stelle sicher, dass bei der Nutzung internationaler Endpunkte keine personenbezogenen Kundendaten im Prompt enthalten sind.
Fazit & Download
Dieser Workflow beweist: Professionelle KI-Infrastruktur muss nicht teuer sein. Mit den richtigen Open-Source-Bausteinen baust du in unter 15 Minuten ein Setup auf, das kommerziellen Agenten-Plattformen in nichts nachsteht.
Original-Inspiration von The AI Leverage: https://theaileverage.beehiiv.com/p/gemini-4-argon-is-now-open-to-paying-users
Du willst mehr praxiserprobte KI-Strategien für dein Business?
Abonniere den kostenlosen ki-netic Newsletter auf ki-netic.ch oder vernetze dich mit uns auf Instagram @ki_netic_schweiz.
Gefällt dir dieser Leitfaden?
Wir schicken dir jede Woche exakt einen praxiserprobten Workflow, der dir im Geschäftsalltag Stunden spart. Kein Spam. Direkt umsetzbar.
Bereits von über 1'200 Freelancern & KMU-Leitern im DACH-Raum gelesen.
