ThinkingNews
Top Stories
7 outlets·9 reports·over 13h

Google unveils Gemini 4 Argon, and cyber defenders get it first

ThinkingNews Desk · how this was written

Google has released the Gemini 4 Argon AI model through its Fairwind Program to a select group of trusted cyber-defenders. The frontier model offers a 1-million-token output limit, outperforms GPT-6 Astra on coding and knowledge benchmarks, reduces hallucinations to about 15%, saves roughly 300 TiB of memory, and can migrate thousands of C/C++ lines to Rust.

Written from all 7 reports below, not from any single one.

How it was reported

  1. Hacker News·
    Gemini 4 Argon

    Gemini 4 Argon is a frontier model released to trusted cyber defenders via Google’s Fairwind Program. It supports a 1 million-token output limit and charges $2 per million input tokens and $10 per million output tokens. The model accelerates quantum algorithmic optimization by 40%, frees 300 TiB of memory across data centers, and migrates up to 800 k lines of C/C++ to Rust with 2.7× speed gains for libgav1. Gemini 4 Argon scores 77.9% on DeepSWE v1.1 and leads the Vals Index for finance, coding, and legal tasks.

  2. TechMeme·
    Google rolls out Gemini 4 Argon to a small group of cybersecurity partners and says it outperforms GPT-6 Astra on certain coding and knowledge work benchmarks (Madison Mills/Axios)

    Google has released its Gemini 4 Argon AI model to select cybersecurity partners through the Fairwind platform. The company claims the model outperforms GPT-6 Astra on specific coding and knowledge benchmarks, though some employees report performance inconsistencies in real-world coding tasks. The model is currently participating in the United States government's voluntary pre-release safety process.

  3. Ars Technica·
    Google announces Gemini 4 Argon AI model, but you can't use it yet

    Google released the Gemini 4 Argon model, claiming industry-leading performance in coding, knowledge work, and cybersecurity. Argon is used internally to process fleet-wide telemetry, saving 300 TiB of memory, and to migrate thousands of C/C++ lines to Rust in libraries such as re2, libgav1, and the Fuchsia OS Zircon kernel. Benchmark results show Argon scoring 77.9% on DeepSWE v1.1, surpassing GPT-6 Astra, Fable 5.1, and Opus 5.5, and achieving top marks on the Vals Index economic analysis test.

  4. The Next Web·
    Google unveils Gemini 4 Argon, and cyber defenders get it first

    Google released Gemini 4 Argon, a frontier AI model featuring a 1-million-token output limit and superior performance on software engineering benchmarks. Cybersecurity defenders currently access the model without safety guardrails to identify vulnerabilities, while paid API and Ultra subscribers will gain access following further testing.

  5. The Verge·
    Google announces Gemini 4 and says it’s so capable that only ‘trusted cyber defenders’ can have it right now

    Google unveiled Gemini 4 Argon, a new AI model that delivers frontier performance in complex workflows for software engineering, enterprise knowledge work such as legal and finance, and cybersecurity defense. The company is initially restricting access to a set of trusted cyber defenders and is participating in the U.S. government’s voluntary pre-release model access program while planning gradual expansion.

  6. Hacker News·
    Gemini 4 Argon (High): Intelligence, Performance and Price Analysis
  7. TechCrunch·
    Google releases Gemini 4 Argon, called its most powerful model yet

    Google launched Gemini 4 Argon, an AI model designed to autonomously identify, validate, and patch software vulnerabilities. Currently available through the Fairwind Program, the model also supports coding, debugging, and visual analysis. According to Vals benchmarking data, Argon outperforms OpenAI’s GPT-6 Astra and Anthropic’s Fable and Opus models across various technical tasks.

  8. TechMeme·
    Artificial Analysis says Gemini 4 Argon (high) matches GPT-6 Astra (max) on its Intelligence Index and has a 15% hallucination rate, compared with 51% for Astra (Artificial Analysis)

    Google has released its Gemini 4 Argon AI model to cybersecurity partners and is participating in US government pre-release safety processes. Artificial Analysis reports that the model matches GPT-6 Astra on its Intelligence Index while reducing hallucination rates to 15% and lowering task costs by 40%.

  9. Engadget·
    Google's first Gemini 4 model is 'Argon'

Related stories

Share: