Vibitely Daily

Thursday, 19 November 2026, story 1 of 7

Google releases Gemini 4 Neon at $3/$15 per million tokens

Neon has a 1 million-token context window and is rolling out in AI Studio, Antigravity and the Gemini app; the benchmark claims are Google's.

Illustration: A glowing orange neon tube standing beside a row of rotary dials, one dial turned halfway, against a plain dark background.

What happened

Google DeepMind introduced Gemini 4 Neon on Nov 18, a model it says is built for software engineering, knowledge work and science. Google prices it at $3 per million input tokens, $0.15 per million cached input tokens and $15 per million output tokens, with a 1 million-token context window.

It is rolling out in the Gemini API in Google AI Studio, Gemini for business, Google Antigravity for paid Google AI Pro and Ultra subscribers, and the Gemini app for Pro and Ultra subscribers. A cyber variant is due for trusted defenders "in the coming days" under the Fairwind Program.

What it means

The headline price is only part of the cost picture. Google says that at the default high thinking level Neon costs 32% less per task than Gemini 4 Carbon. That is Google's figure, and the source gives no per-token prices for other Gemini 4 models, so a per-token comparison is not possible from it. Cost per task depends on how many tokens a model spends, so teams should measure it on their own workloads.

The coding numbers carry a caveat Google supplies itself. On SWE-sweep, Neon scores 12.8% on the mini-SWE harness and 39.0% in Google Antigravity. Results may depend heavily on the tooling around the model, and teams outside Antigravity could see less. The source also does not say whether this is a preview or general availability.

What to do: Run Neon at its default high level on your own tasks and compare cost per task with your current model before trusting Google's benchmark figures.

Sources

Filed under Models. Published .

"What it means" is analysis, not reporting. Found an error? Write to daily@vibitely.com.