# Google Antigravity — Gemini 4 Neon in Google Antigravity

- Company: Google Antigravity (antigravity.google)
- Announced: 2026-11-18T08:00:00+00:00
- Category: availability-change
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://antigravity.google/blog/gemini-4-neon-in-google-antigravity
- Record: https://forck.live/items/38269-gemini-4-neon-in-google-antigravity
- Subject: Antigravity
- Models affected: Gemini 4 Neon, Gemini 4 Carbon, Gemini 4 Hydrogen

Google is rolling out Gemini 4 Neon inside Google Antigravity for paid Google AI Pro and Google AI Ultra subscribers, positioning it between Gemini 4 Carbon and Gemini 4 Hydrogen. Google says Neon is built for the middle of daily coding and knowledge work: complex enough to need frontier-level reasoning, frequent enough to demand speed on every turn. Neon ranks #3 overall with a score of 61 on the Artificial Analysis Intelligence Index at its Extra High thinking level, and ranks #1 on the Artificial Analysis Coding Agent Index (58.6) and Zapier's AutomationBench (56.3%), according to the company. At its default High thinking level, Google says Neon shows a 32% lower cost per task than Carbon, with four thinking levels from Low to Extra High. Orchestrator in Google Antigravity now supports pairing Neon as planner with Hydrogen as executor in one session, handing routine edits, package updates and test runs to Hydrogen and returning to Neon when Hydrogen hits friction. Starting today, Neon is rolling out in the Antigravity desktop app, Antigravity CLI and Antigravity IDE Extensions, and enterprises can access it through Gemini for business.

## Evidence

Verbatim from https://antigravity.google/blog/gemini-4-neon-in-google-antigravity:

> Today, we’re rolling out Gemini 4 Neon inside Google Antigravity. Bringing frontier-level intelligence with fast end-to-end task completion, Neon hits the sweet spot between Gemini 4 Carbon for your deepest long-horizon projects and Gemini 4 Hydrogen for rapid everyday execution.
> 
> Built for demanding daily workflows
> 
> Most coding and knowledge work sits right in the middle: complex enough to need frontier-level reasoning, yet frequent enough to demand speed and efficiency on every turn. Gemini 4 Neon is engineered for real-world AI coding and agent harnesses, handling repo-level edits, debugging, refactoring, secure code generation, grounded enterprise analysis, and STEM reasoning with reliable multi-step instruction following and our lowest fabrication rate yet in internal Google Antigravity testing.
> 
> Frontier-level coding and agentic performance: Neon ranks #3 overall with a score of 61 on the Artificial Analysis Intelligence Index at its Extra High thinking level and ranks #1 overall on the Artificial Analysis Coding Agent Index (58.6) and Zapier’s AutomationBench (56.3%).

---

Record: https://forck.live/items/38269-gemini-4-neon-in-google-antigravity
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md
