Official Google blog post Introducing Gemini 4 Neon, dated 18 November 2026, by Koray Kavukcuoglu, with an orange neon-style Gemini 4 Neon sign
Official Introducing Gemini 4 Neon announcement on blog.google, captured 19 November 2026. Image: AppFor.ai screenshot of blog.google. Image source ↗

What Google announced

On 18 November 2026, Koray Kavukcuoglu of Google DeepMind published Introducing Gemini 4 Neon. The post presents Neon as a model that aims for frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work, and science workflows. Google places it in the Gemini 4 family between everyday Hydrogen work and Carbon's hardest long-horizon projects: Neon, it says, balances depth and efficiency for complex coding, analytical reasoning, and production-grade enterprise workflows.

At its xhigh thinking level, Google reports that Neon ranks ahead of comparable models and number 3 overall, with a score of 61, on the Artificial Analysis Intelligence Index, just behind larger, slower models. It also reports number 1 overall on the Artificial Analysis Coding Agent Index (58.6), 65.2% on Terminal-Bench 4.1, and number 2 on the Vals Index (70.9%). Those figures are Google's published claims, not independent measurements in this article.

Who can use it, and what it costs

Google says Gemini 4 Neon is rolling out in the Gemini API in Google AI Studio, in Gemini for business via the API and model dropdown, for paid Google AI Pro and Ultra subscribers in Google Antigravity, and for Pro and Ultra subscribers in the Gemini app. The official model page identifies the API model as gemini-4-neon. The live Gemini Developer API pricing page, checked on 19 November 2026, states that Neon is not available on the Free tier.

The announcement and the live pricing page both list $3 per million input tokens and $15 per million output tokens. The pricing page also lists a context caching price of $0.15 per million tokens, with implicit caching only, which matches the announcement's $0.15 cached input figure (a 95% discount). Output prices on the official pricing page include thinking tokens. Neon has a 1 million-token context window (1,048,576 tokens on the model page).

How Google compares it with other Gemini 4 models

Official docs say Neon supports four thinking levels: low, medium, high (the default), and xhigh. Carbon and Argon default to medium and cap at high. At default high, Google says Neon pairs a high Artificial Analysis Intelligence Index score with a 32% lower cost per task than Carbon. That is a cost-per-task claim, not a claim that Neon's list token price is lower than Carbon's current promotional Standard rates of $2 input and $10 output per million tokens through 30 November 2026 (then $4 and $20 from 1 December 2026 on the same pricing page). A Gemini 4 Neon for cyber variant is described as coming to Fairwind Program members in the coming days, not as already generally available.

Sources & further reading

Based on the official documentation below. Features and access can change; check the provider before making a decision.

  1. Google: Introducing Gemini 4 Neon, dated 18 November 2026, retrieved 19 November 2026 (opens in a new tab)
  2. Official Gemini Developer API pricing page (ai.google.dev/gemini-api/docs/pricing), rendered 19 November 2026 (opens in a new tab)
  3. Official Gemini 4 Neon model page, rendered 19 November 2026 (opens in a new tab)
  4. Official What's new in Gemini 4 Neon page, rendered 19 November 2026 (opens in a new tab)
Something out of date or incorrect?Report a correction

Was this page useful?

Help us make our coverage more useful. This is feedback on the page, not a rating of the tool.

Share this page

Related reading