By the toolist.ai editors. Researched from public sources, not hands-on tested by toolist.ai. Updated November 19, 2026; prices checked November 19, 2026.
Gemini 4 Neon: price, context window and benchmarks
Gemini 4 Neon is the Gemini 4 model Google places between Gemini 4 Hydrogen and Gemini 4 Carbon, for complex coding, knowledge work and science. It costs $3 per million input tokens and $15 per million output tokens with no introductory price, against Carbon's introductory $2 and $10 that rise to $4 and $20 on December 1, 2026, and Artificial Analysis puts it at 61 on its Intelligence Index at xhigh thinking and 60 at high, the API default.
Best for
- Complex coding, knowledge work and science tasks that need more depth than Gemini 4 Hydrogen, where Google says Neon finishes faster and with fewer output tokens than Gemini 4 Carbon
- Hard multi-step problems at xhigh thinking: 64% on Terminal-Bench 4.0, 71% on HLE-Diamond and 70% on Terminal-Bench-Science 0.1, according to Artificial Analysis
- Long prompts at one price: $3 input and $15 output per million tokens across the whole 1,048,576-token window, with cached input at $0.15 per million
- Very long answers: with long decode (background execution), one response can use up to 1,048,576 tokens minus the input, past the 65,536-token default
Key facts
| Fact | Gemini 4 Neon | Source |
|---|---|---|
| Provider | Gemini API docs: Pricing | |
| API model name | gemini-4-neon | Gemini API docs: Gemini 4 Neon |
| Released | November 18, 2026 | Gemini API docs: Release notes |
| Input price | $3 per million tokens | Gemini API docs: Pricing |
| Cached input price | $0.15 per million tokens | Gemini API docs: Pricing |
| Output price | $15 per million tokens | Gemini API docs: Pricing |
| Context window | 1,048,576 tokens | Gemini API docs: Gemini 4 Neon |
| Maximum output | 65,536 tokens | Gemini API docs: Gemini 4 Neon |
| Knowledge cutoff | March 2026 | Google DeepMind: Gemini 4 Neon model card |
| Input and output | Text, images, video, audio and PDF in; text out | Gemini API docs: Gemini 4 Neon |
| Reasoning | Thinking is on by default. Thinking levels low, medium, high and xhigh; high is the API default | Gemini API docs: Gemini thinking |
| API and cloud platforms | Gemini API paid tier (in Google AI Studio and Android Studio), Gemini for business through the API, and Google Antigravity for businesses and for paid Google AI Pro and Ultra subscribers. Eligible members of the Fairwind Program, for cyber defense, get access over the days after launch | Google: Introducing Gemini 4 Neon |
| In the apps | Gemini app and Google AI Studio on Google AI Pro and Ultra; Gemini for business through its model drop-down | Google: Introducing Gemini 4 Neon |
| Data retention | On the paid tier, Google does not use prompts or responses to improve its products; it logs them for a limited period only to detect misuse and for legal disclosures | Gemini API Additional Terms of Service |
Output includes thinking tokens. Caching is implicit only; explicit caching is not supported. There is no long-context surcharge. Batch and Flex cost half; Priority costs 1.8 times the standard price. Grounding with Google Search and with Google Maps each has 5,000 free requests a month shared across Gemini 3 and newer models, then $14 per 1,000. Not available on the free tier.
What Gemini 4 Neon costs a month
Three typical workloads, priced at the standard API rates above. We calculated these; the provider does not publish them.
| Workload | Tokens a month | Cost a month |
|---|---|---|
| Support chatbot 100,000 replies, each 3,000 input and 400 output tokens, no caching | 300M in, 40M out | $1,500 |
| Document Q&A 5,000 questions over the same 60,000-token document set (90% cache hits), 800 output tokens each | 300M in, 4M out | $190 |
| Coding agent 200 tasks, each 1.5 million input tokens (90% cache hits) and 50,000 output tokens | 300M in, 10M out | $281 |
Reasoning models bill their thinking as output tokens, so heavy reasoning can raise the output count several times. Cache-write charges are not included.
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| Artificial Analysis Intelligence Index xhigh thinking, the highest level Artificial Analysis tests | 61.0 | Artificial Analysis: Gemini 4 Neon |
| Intelligence Index at everyday effort high thinking, the API default | 60.0 | Artificial Analysis: Gemini 4 Neon |
| Terminal-Bench 4.0 xhigh thinking | 64.0% | Artificial Analysis: Gemini 4 Neon |
| HLE-Diamond xhigh thinking | 71.0% | Artificial Analysis: Gemini 4 Neon |
| Terminal-Bench-Science 0.1 xhigh thinking | 70.0% | Artificial Analysis: Gemini 4 Neon |
Speed: about 101.5 output tokens a second (high thinking, Artificial Analysis; Artificial Analysis: Gemini 4 Neon).
Benchmark and speed data marked as such come from Artificial Analysis.
What changed
Compared with Gemini 4 Carbon, which Google calls its strongest reasoning model, Google says Neon completes tasks faster and spends fewer output tokens, and it places Neon between Carbon and Gemini 4 Hydrogen, with harder reasoning than Hydrogen. Neon's $3 input and $15 output per million tokens are above Carbon's introductory $2 and $10, which run through November 30, 2026, and below Carbon's $4 and $20 from December 1; Neon has no introductory price, and Hydrogen costs $1.50 and $7.50. Neon defaults to high thinking and adds xhigh, where Carbon defaults to medium and stops at high, and Google rates Neon's speed moderate against Carbon's deliberate. Neither is on the free tier. Google recommends Neon as the upgrade from Gemini 3.1 Pro Preview. (Gemini API docs: Choosing a model)
Gemini 4 Neon compared
Other models
- GPT-6.2 Astra: $8 in, $40 out per million tokens
- GPT-6.2 Sol: $2 in, $10 out per million tokens
- Gemini 4 Carbon: $2 in, $10 out per million tokens
- Claude Fable 5.2: $8 in, $40 out per million tokens
- GPT-6.1 Terra: $1 in, $5 out per million tokens
- Gemini 4 Argon: $2 in, $10 out per million tokens
- Muse Wonder: $4 in, $20 out per million tokens
- Claude Haiku 5.5: $0.10 in, $0.50 out per million tokens
- GPT-6.1 Sol: $2 in, $10 out per million tokens
- Claude Sonnet 5.5: $2 in, $10 out per million tokens
- Claude Opus 5.5: $4 in, $20 out per million tokens
- GPT-6 Luna: $0.10 in, $0.50 out per million tokens
- GPT-6 Sol: $2 in, $10 out per million tokens
- GPT-6 Astra: $10 in, $50 out per million tokens
- Claude Fable 5.1: $10 in, $50 out per million tokens
- GLM-5.3: $1.40 in, $4.40 out per million tokens
- Claude Opus 5: $5 in, $25 out per million tokens
All prices side by side: AI model API price table. How we research: our method.
Sources
- Gemini API docs: Gemini 4 Neon (checked 2026-11-19)
- Gemini API docs: Pricing (checked 2026-11-19)
- Gemini API docs: Release notes (checked 2026-11-19)
- Gemini API docs: Gemini thinking (checked 2026-11-19)
- Gemini API docs: Choosing a model (checked 2026-11-19)
- Gemini API docs: What's new in Gemini 4 Neon (checked 2026-11-19)
- Gemini API docs: Long decode (checked 2026-11-19)
- Google DeepMind: Gemini 4 Neon model card (checked 2026-11-19)
- Google: Introducing Gemini 4 Neon (checked 2026-11-19)
- Gemini API Additional Terms of Service (checked 2026-11-19)
- Artificial Analysis: Gemini 4 Neon (checked 2026-11-19)
- OpenRouter: Gemini 4 Neon (checked 2026-11-19)