Bookmarks

    Nothing saved yet. Use the bookmark on any article. Bookmarks are kept on this device only: they stay private and work without an account.

    Settings

    Appearance

    Reading mode

    Font size

    Content width

    Line spacing

    Keyboard shortcuts

    /
    Search
    j k
    Next / previous story
    ← →
    Previous / next section (columns)
    b
    Bookmarks
    s
    Settings
    r
    Columns / scroll
    t
    Light / dark

    Saved in this browser only.

    Ping2Prompt
    Subscribe to the Weekly Brief Sign in
    Illustration for: Google releases Gemini 4 Neon, a fast reasoning model with a 1M-token context
    Photo by Solen Feyissa / Unsplash

    Google releases Gemini 4 Neon, a fast reasoning model with a 1M-token context

    Google AI

    Google adds Gemini 4 Neon to its model lineup

    Google has released Gemini 4 Neon, a model it aims at software development, knowledge work and science. Google's listing says Neon builds on Gemini 4 Hydrogen with deeper reasoning.

    The Decoder counts Neon as the fourth new Gemini model in a month, and OfficeChai puts the four Gemini 4 models at about seven weeks. Per 36Kr, Argon was announced September 30, Carbon on October 28 and Hydrogen on November 12.

    Neon takes five input types and a one-million-token window

    • Model ID: the identifier is gemini-4-neon, tagged for generation, extraction and recognition tasks.
    • SDKs: the Vertex AI listing says you can reach it through Google's Gen AI SDK in Python, Go, JavaScript, Java and .NET.

    Google's "close to Carbon" claim has no scores behind it

    Google's Vertex AI listing describes Neon as a fast reasoning model with quality near Gemini 4 Carbon, at higher speed and lower cost. That is Google's own characterisation, not a measured result, and no Artificial Analysis or other independent score for Neon is given. For reference, per OfficeChai, the Artificial Analysis Intelligence Index scored Hydrogen at 54 at its highest reasoning setting and Carbon at 59 at launch.

    Neon pricing sits at $3 and $15 per million tokens

    Neon costs $3 per million input tokens and $15 per million output tokens, and OfficeChai says Google's Cloud SKU catalog already listed those rates. The Decoder says that is above earlier Gemini 4 models. Hydrogen costs $1.50 input and $7.50 output per million tokens, per OfficeChai. Carbon's introductory $2 and $10 is set to become $4 and $20 on December 1, per OfficeChai.

    Enterprises can reserve capacity for Neon through Provisioned Throughput. OfficeChai points to Anthropic's Claude Sonnet 5.5, at $2 and $10 per million tokens, as a possible rival.

    Test Neon on your own workload before switching

    If you pay per token, price Neon against Hydrogen and Carbon using your own prompts, since Google's quality claim comes without scores. Carbon's scheduled price change is worth including in that comparison.

    What Google has not published about Neon

    No rate limits or regional availability are given. No exact public release date is given. No Neon-specific deprecation or migration details are given.

    Sources


    This story was produced with AI assistance from the sources listed above and fact-checked against them before publication. Read our editorial methodology and the corrections log. Spotted an error? Email [email protected].

    Continue

    Share on Bluesky · Mastodon · Email ·

    Discussion