Skip to content
Brief The AI-First Web ·

Google releases Gemini 4 Neon, its fourth new Gemini 4 model in four weeks

Neon costs more than two sister models until December 1, then less.

In brief
  • Neon costs $3 per million input tokens and $15 per million output tokens. That is above Carbon and Argon until December 1, then below them.
  • Benchmark results were not out yet. The Decoder said they should follow within hours.

The Decoder reported on 18 November that Google has released Gemini 4 Neon, a new AI model. Google says it brings frontier-level intelligence and fast task completion to software engineering, knowledge work and science. It is available now through Google's API. Google charges $3 for every million tokens sent in and $15 for every million tokens returned. Tokens are the small chunks of text a model reads and writes.

The Decoder lists Carbon and Argon, two other Gemini 4 models, at an introductory $2 and $10, rising to $4 and $20 on December 1. So Neon costs more than they do until then, and less after. Hydrogen is priced at $1.50 and $7.50. The Decoder does not say whether the December prices will last. Neon's context window, the amount of text it can weigh at once, is about one million tokens, and its replies can reach one million. It takes text, images, video, audio and PDFs, but cannot generate images or audio. Benchmark results were not out yet.

For people who budget for AI, price alone does not rank these models, because the order flips on December 1. Four releases in four weeks, with prices that change, make spending hard to plan. Teams should check which tier each job needs and whether their cost estimates still hold after that date.

A WebPulse Brief: a short report of an important event, written by the WebPulse Newsroom with AI assistance and checked against the reporting below. When there is more to explain, we follow up with a full story. How we use AI.

Reporting: The Decoder.