# Gemini 4 Neon

> Google's mid-tier Gemini 4 model, announced November 2026 for long-running coding, knowledge work and science, with a 1M-token context window.

- Maker: Google
- Line: Gemini 4
- Canonical page: https://ai-tldr.dev/models/gemini-4-neon/
- Updated: 2026-11-18

## In short

Gemini 4 Neon is the model Google announced on 18 November 2026, in a post by Koray Kavukcuoglu. Google says it targets frontier-level intelligence with fast task completion in software engineering, knowledge work and science. It costs $3 per million input tokens and $15 per million output tokens.

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Input price | $3 / 1M tokens | [1] |
| Output price | $15 / 1M tokens | [1] |
| Licence | Proprietary · API only | [2] |
| Released | 18 November 2026 | [2] |
| Parameters | Undisclosed | [2] |

## Sources

1. [Google — Introducing Gemini 4 Neon](https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-neon/)
2. [Google DeepMind — Gemini 4 Neon model card](https://storage.googleapis.com/deepmind-media/Model-Cards/Gemini-4-Neon-Model-Card.pdf)

## Overview

Gemini 4 Neon is the model Google announced on 18 November 2026, in a post by Koray Kavukcuoglu. Google says it targets frontier-level intelligence with fast task completion in software engineering, knowledge work and science.

It sits between Gemini 4 Carbon, which Google keeps for the hardest long-horizon projects, and Gemini 4 Hydrogen, which handles fast everyday execution. It has a 1M-token context window.

Google designed it for long-running, tool-heavy agent workflows, including as an Orchestrator that directs parallel Hydrogen subagents. A cyber version is coming to Fairwind Program defenders over the following days.

## Benchmarks

- Terminal-Bench 4.1: 65.2% ([source](https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-neon/))
- HLE-Diamond: 71.3% ([source](https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-neon/))
- Agents' Last Exam v1.1: 75.8% ([source](https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-neon/))

## Pricing

- Input: $3 / 1M tokens
- Output: $15 / 1M tokens

Cached input costs $0.15 per 1M tokens.

Source: https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-neon/

## Strengths

- 61 on the Artificial Analysis Intelligence Index at xhigh thinking, ranked #3 overall
- 58.6 on the Artificial Analysis Coding Agent Index, ranked #1
- 65.2% on Terminal-Bench 4.1
- 32% lower cost per task than Gemini 4 Carbon at high thinking, per Google

## Best for

- Reach for it for long-running coding agents and tool-heavy workflows
- Reach for it as an Orchestrator that directs Gemini 4 Hydrogen subagents
- Reach for it for legal, tax, finance and spreadsheet work, where Google reports #1 results on several benchmarks
- For the hardest long-horizon projects Google points to Gemini 4 Carbon instead

## Gemini 4 versions

- Gemini 4 Neon (18 November 2026): Mid-tier Gemini 4 model for long-running coding and agent work. — this page

## FAQ

### How much does Gemini 4 Neon cost?

Gemini 4 Neon costs $3 per million input tokens and $15 per million output tokens, according to Google. Cached input tokens cost $0.15 per million, a 95% discount.

### Where can I use Gemini 4 Neon?

Gemini 4 Neon is rolling out from 18 November 2026 in the Gemini API, Google AI Studio and Android Studio, in Gemini for business, in Google Antigravity for paid AI Pro and Ultra subscribers, and in the Gemini app for Pro and Ultra subscribers.

### How does Gemini 4 Neon compare with Gemini 4 Carbon?

Google says Gemini 4 Neon costs 32% less per task than Gemini 4 Carbon at high thinking. Carbon stays Google's choice for the hardest long-horizon projects, while Neon targets fast task completion in coding, knowledge work and science.

## Links

- [Google — Introducing Gemini 4 Neon](https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-neon/)
- [Google DeepMind — Gemini 4 Neon model card](https://storage.googleapis.com/deepmind-media/Model-Cards/Gemini-4-Neon-Model-Card.pdf)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/gemini-4-neon/
