Gemini 4 Neon is now available to Google AI Pro and Ultra subscribers: open the model dropdown in the Gemini app and switch to it. The free plan cannot use it yet. In its official blog post on November 18, 2026, Google DeepMind introduced Neon as a new member of the Gemini 4 family, built for complex software engineering, knowledge work, and scientific tasks. It scores 61 on the Intelligence Index, ranking third, with API pricing of $3 per million input tokens and $15 per million output tokens, and a 1-million-token context window.
Switching to Neon in the Gemini App
Open the Gemini app, tap the model selector at the top, find Gemini 4 Neon, and tap it to switch; you can then use it in that conversation for long documents, spreadsheets, charts, and code. Both Pro and Ultra subscribers can select it: the difference is mainly the usage allowance, not access. If you do not see the option yet, the rollout probably has not reached your account: first confirm your subscription is active and update the app to the latest version, then check again a day later. Paid Pro and Ultra subscribers can also select Neon in Google Antigravity, and developers can access it through the paid API in Google AI Studio.
What the API Costs
The Neon API costs $3 per million input tokens and $15 per million output tokens; cached input costs only $0.15, a 95% discount. The 1-million-token context window suits feeding in large amounts of material and then following up over several turns. Google also offers four thinking levels: at the default high level, it says each task costs about 32% less than Carbon. For comparison, Carbon's introductory pricing also ends after November 30, after which Neon becomes the cheaper option. There is no need to use Neon for everyday chat or short questions: it is built for long tasks and complex reasoning, while lighter needs are better served by the Hydrogen tier.
Free Plan or Pro: How to Choose
The free plan currently has no access to Neon. The cheapest way to use it in the Gemini app is a Google AI Pro subscription; Ultra mainly adds higher usage allowances and benefits in other products, not exclusive access to Neon itself. If you only need it once or twice for a long task, paying per use through the API is more economical: you pay only for what you use. For the full pricing and positioning from launch, compare our Gemini 4 Neon launch report; if you are also weighing Pro against Ultra, see the Genie 4 launch report from the same period, since the two features have different subscription requirements.