November 18, 2026: Google began rolling out Gemini 4 Neon to Google AI Pro and Ultra subscribers in the Gemini app. Its official announcement promises faster work on tough tasks, such as writing and sorting through complex documents.
Already paying for Gemini? Neon gives you a reason to revisit a tricky request. But Google’s launch claims alone don’t justify changing subscriptions (another monthly bill deserves more than a leaderboard).
What Happened: Gemini 4 Neon Joins the App
Google introduced Neon as a new addition alongside its existing Hydrogen and Carbon models, though its official announcement doesn’t spell out exactly how the three are meant to compare. The launch also includes the Gemini API through Google AI Studio, Gemini for business, and paid Antigravity access.
According to 9to5Google’s report, Neon replaces Argon and Gemini 3.1 Pro in the Gemini app, though Google’s own materials don’t confirm this exact swap. The outlet describes the picker as including Auto, Everyday help: 4 Hydrogen, Deeper analysis: 4 Neon, and Code & complex work: 4 Carbon.
The announced app rollout covers Google AI Pro and Ultra subscribers. Free users shouldn’t count on Neon access based on this announcement.
On Windows
For Windows readers, the change to watch is the Gemini model choice. Look for the reported Deeper analysis option as access reaches your paid account.
For Android
Android readers should watch for the same paid-account rollout. Google says access is rolling out, so subscribers won’t all get it at the same time.
On the Web
Keep the Gemini app and Google AI Studio separate when comparing costs. The app offers subscription access. AI Studio provides access to the API, which charges based on use.
What Could Improve in Everyday Use?
Google says Neon can combine information across multiple file types, including documents, spreadsheets, and other media. Its announced 1 million-token context window is like a larger desk for spreading out material. Tokens are small units of model input, and the app’s upload allowance is a separate limit.
That capacity could help Neon connect details across documents rather than summarize each one on its own. 9to5Google also reports clearer writing and possible uses for utility bills, receipt comparisons, and purchase research.
For example, choosing a mesh Wi-Fi system or external SSD means weighing several needs beyond price. Google positions Neon to help work through those details together. You’ll still need to check the specs and cited prices yourself.
How It Stacks Up
Google’s published results make Neon worth watching, but they don’t give it a clean sweep.
Every figure below is reported by Google on its Gemini comparison page. That includes scores on tests developed outside Google. These aren’t independent consumer review results.
| Benchmark | Gemini 4 Neon | Claude Opus 5.5 | GPT-6.2 Sol |
|---|---|---|---|
| Artificial Analysis Intelligence Index v5.0.1 | 61 | 59 | 57 |
| Vals Index v2.2 | 70.9% | 70.1% | 66.1% |
| Chartography | 73.5% | 66.3% | 75.1% |
| Terminal-Bench 4.1 | 65.2% | 65.8% | 61.5% |
Google places Neon third overall on the Artificial Analysis index at its highest thinking setting. Its 61 is an index score, not a percentage. Google places it second overall on Vals.
GPT-6.2 Sol leads this selection on chart understanding. Claude Opus 5.5 edges Neon on Terminal-Bench, a test of technical tasks. These comparisons don’t show that each model used equal settings. They also won’t tell you how well each will explain your electricity bill.
What Does “Faster” Mean?
Google’s speed claim covers how quickly Neon finishes complex tasks, not a head-to-head timing comparison against other models. It doesn’t give measured response times for the Gemini app or independent test results for everyday reliability.
Google also says Neon produces fewer inaccurate or invented details than its earlier models, based on its own internal testing. That’s a claim about Google’s own tests, not an independent measurement. You’ll still need to check key details in documents.
Google describes four thinking levels that balance reasoning effort and cost. Don’t assume the app gives you four matching switches. The reported choice for app users is Deeper analysis.
Subscription Access Versus API Pricing
Google’s November 18 announcement lists these API charges:
- Input: $3 per million tokens.
- Cached input: $0.15 per million tokens.
- Output: $15 per million tokens.
Those are usage charges for software calling the model. They don’t tell you the subscription price or the cost of a single Gemini app conversation.
The Reaction
Independent reaction is still forming across X, Reddit, and Hacker News. Google’s own posts don’t show a shared user verdict on speed or reliability.
Our Take
Existing Pro and Ultra subscribers have the clearest reason to try Neon. Long documents and questions with several requirements are sensible places to start. Hydrogen remains Google’s everyday-help choice. A simple question doesn’t automatically need deeper analysis.
Wrapping Up
When Neon appears, give it one familiar document task and check its answer against the original. Google’s benchmarks justify a trial, but I’d base any subscription change on results from your own work. Save the prompt so you can repeat the same comparison with your current assistant.