Google Expands Gemini Family With New Mid Tier Model
Google LLC released Gemini 4 Neon on November 18, 2026, creating a new middle tier in its artificial intelligence lineup that sits between the deep engineering focus of Gemini 4 Carbon and the fast everyday execution of Gemini 4 Hydrogen. The company announced the new model to deliver advanced reasoning and multi-step task handling without requiring users to wait for heavier models to deliberate. By replacing the retired Gemini 4 Argon model in the application model picker, Google is restructuring its fourth generation offerings around four distinct operational levels. The release brings faster response times and lower costs to complex tasks such as software engineering, financial modeling, and scientific research.
Contents
- Google Expands Gemini Family With New Mid Tier Model
- How Neon Compares on Benchmarks and Speed
- Pricing Structure and the November Promotional Window
- Orchestrator Mode and Multi Agent Workflows
- Availability Across Consumer and Enterprise Channels
- Cybersecurity Variant and Specialized Testing
- A Rapid Succession of Fourth Generation Releases
- At a Glance
How Neon Compares on Benchmarks and Speed
Independent evaluation by Artificial Analysis places Gemini 4 Neon at number three overall on the Artificial Analysis Intelligence Index with a score of 61 when operating at its highest thinking level. Google reported that the model scored 58.6 on the Artificial Analysis Coding Agent Index, earning the top spot in that evaluation. Koray Kavukcuoglu, senior vice president of Google DeepMind and Google's chief AI architect, announced the model on November 18, 2026, stating that it delivers frontier-level intelligence with fast end-to-end task completion.
To me Neon feels like the right mix of intelligence, speed and usability. It is absolutely punching well above its model class in a meaningful way.
That quote comes from Logan Kilpatrick in a public post discussing the capabilities of the new model. Google stated that Neon completes tasks about three times as quickly as Carbon while cutting per-task costs by roughly 32% to 60% depending on the benchmark workload. The model supports a context window of up to one million tokens in standard developer environments and up to 512,000 tokens inside Google Antigravity, allowing users to process dense spreadsheets, PDFs, audio, and video files within a single prompt.
Pricing Structure and the November Promotional Window
Developers accessing the model through the paid Gemini API pay $3.00 per million input tokens and $15.00 per million output tokens, with cached input tokens priced at $0.15 per million. Because the flagship Gemini 4 Carbon model is currently running on an introductory promotional rate of $2.00 per million input tokens and $10.00 per million output tokens through November 30, 2026, Neon appears temporarily more expensive than its heavier sibling for developers. Once Carbon's introductory pricing period ends on December 1, 2026, its standard rate rises to $4.00 per million input tokens and $20.00 per million output tokens, making Neon the more cost-effective mid tier option per token.
Orchestrator Mode and Multi Agent Workflows
Google Antigravity users can access a specialized feature called Orchestrator Neon, which pairs two separate models within a single conversation thread to handle complex programming tasks. In this setup, Neon acts as the lead architect to plan and review code modifications, while Gemini 4 Hydrogen executes the implementation work such as file edits, package updates, and test runs. If Hydrogen encounters an error or gets stuck during execution, control automatically returns to Neon to inspect the system state, correct the plan, and continue the workflow. Developers can also adjust Neon's reasoning effort across four distinct thinking levels named low, medium, high, and extra high, trading processing time against token expenditure.
Availability Across Consumer and Enterprise Channels
Paid subscribers of Google AI Pro at $19.99 per month and Google AI Ultra starting at $99.99 per month can access Gemini 4 Neon immediately inside the Gemini app by selecting the deeper analysis option in the model dropdown. Developers can integrate the model into custom applications using Google AI Studio and Android Studio, while enterprise teams can access it through Gemini for Business. Free tier users do not receive access to the new model, as free accounts continue running on Auto, which automatically assigns a fast lightweight model such as Hydrogen for each prompt.
Cybersecurity Variant and Specialized Testing
Google is also preparing a dedicated security build named Gemini 4 Neon for Cyber, which will reach vetted defenders in the coming days through the company's Fairwind Program. Testing data published by Google indicates that the cybersecurity variant achieves strong scores in finding software vulnerabilities and generating secure code patches, following the precedent set by earlier restricted releases such as the one detailed in Google unveils Gemini 4 Carbon. These specialized tools build on the company's ongoing safety evaluations while providing automated assistance for defensive engineering tasks.
A Rapid Succession of Fourth Generation Releases
The introduction of Gemini 4 Neon continues a fast release cadence that saw Google deploy multiple family members in quick succession following the rollout of Gemini 4 Carbon on November 10 and Gemini 4 Hydrogen on November 12, 2026. This rapid expansion mirrors other recent ecosystem changes, such as the hardware-focused integrations seen in Gentle Monster Unveils Gemini AI Glasses and the broader rollout of assistant capabilities highlighted in Google makes Gemini Tasks free. By constantly updating its model roster, Google aims to provide specific tools for every computational bottleneck without sacrificing system reliability.
At a Glance
- Gemini 4 Neon launched on November 18, 2026, as a mid tier AI model bridging deep reasoning and fast execution.
- The model achieved a score of 61 on the Artificial Analysis Intelligence Index and 58.6 on the Coding Agent Index.
- Pricing is set at $3.00 per million input tokens and $15.00 per million output tokens for API users.
- Paid Google AI Pro and Ultra subscribers can access Neon directly in the Gemini app and Google Antigravity.
- Orchestrator Neon pairs Neon for planning with Hydrogen for fast code implementation.






