← Back
Google DeepMind

Gemini 4 Argon: Built to solve your hardest problems

•5 min read•
#llm#deployment
Level:Advanced
For:AI Engineers
✦TL;DR

Gemini 4 Argon is now globally available via API in AI Studio, Gemini platform, Google Antigravity, and the Gemini app, priced at $2 per M input tokens and $10 per M output tokens (cached input tokens 95 % off). These results demonstrate Argon’s ability to accelerate complex, multi‑step tasks across coding, finance, legal, cybersecurity, and scientific domains. Engineers can immediately leverage Argon through the listed Google services to cut development cycles and improve performance‑critical workloads.

⚡ Key Takeaways

  • Introductory pricing: $2 per M input tokens, $10 per M output tokens, cached input tokens 95 % off.
  • Argon achieved #1 ranking on the Vals Index, a benchmark across finance, coding, legal, and tax work.
  • Agentic code submissions increased by >35 % while code rollbacks dropped by nearly a third.
  • In‑memory session preparation time cut by 40 % and cold import time improved by 21 % for the ADK Python codebase.
  • Quantum chemistry package xTB became up to 2.3× faster at structure prediction and 1.8× faster at conformer search.
💡 Why It Matters

The performance and cost profile lets engineers replace manual optimization loops with Argon‑driven automation, delivering faster releases and higher reliability in production AI systems.

✅ Practical Steps

  1. Call the Gemini 4 Argon model via the API in AI Studio or Google Antigravity using your subscription credentials.
  2. Provide task prompts (e.g., code refactoring goals or .agents/skills rulebooks) to let Argon generate optimizations autonomously.
  3. Integrate the generated changes into your codebase (e.g., update google/adk‑python or contribute patches to xTB) to realize the reported speedups.

Want the full story? Read the original article.

Read on Google DeepMind ↗

More like this

Censored vs. Truncated Regression in Python, Explained

Towards Data Science•#llm

New in AWS Quick: More capable flows, an updated web experience, and agents that work together

AWS ML Blog•#agents

Three insights you may have missed from theCUBE’s coverage of the Dell Technologies New York Symposium

SiliconANGLE AI•#deployment

How Google Duplex evolved to also call businesses for you in Gemini

Google Research•#agents

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING