Summary
Google has made Gemini 4 Carbon available to Google AI Ultra subscribers and to paying developers and enterprises through its API and Gemini for business. The model is designed for complex, multi-step work and uses more tokens than Gemini 4 Argon.
Google announced on November 10, 2026, that Gemini 4 Carbon is available to developers, enterprises, scientists and consumers. The model is rolling out to Google AI Ultra subscribers through Google Antigravity, the Gemini app and AI Mode in Search. Paying developers and enterprises can also use it through the API in Google AI Studio and Gemini for business.
Google positions Carbon as a model for demanding, extended tasks, rather than as a replacement for Gemini 4 Argon across the board. Its announcement describes Argon as the fit for most software engineering and knowledge work, while Carbon is intended for work such as complex coding and debugging and open-ended scientific research.
A model for longer, more complex tasks
Carbon is built for agentic workflows: tasks in which a model can work through multiple steps, use tools and check intermediate results. Google says Carbon uses more tokens than Argon to explore problems, reason through them and run extended verification loops. That distinction matters because the models have the same introductory API price per token, but a Carbon task may consume more tokens.
Google reports that Carbon ranked first on CursorBench 4.0, Perplexity’s WANDR and Zapier’s AutomationBench, at more than twice lower cost per task than the next-best external model on those evaluations. On a separate multimodal agent benchmark, the Multimodal Knowledge Agent (MKA), developed with Artificial Analysis, Google says Carbon ranked second and had more than six times lower average cost per task than the top-performing model. These evaluations cover particular workflows, including work with dense documents and data; the results are not a single general-purpose measure of performance.
Google is also previewing /swarm in Antigravity. It lets a user start a coordinated team of agents from one prompt, assign Carbon roles such as architectural design or auditing, and pair it with faster models working in parallel. Google’s examples include agents building interactive 3D projects from engineering drawings and other source material.
Access and introductory pricing
For paying developers and enterprises using the API or Gemini for business, Google lists the same introductory price as Gemini 4 Argon: $2 per million input tokens and $10 per million output tokens. Google says Carbon is rolling out to all AI Ultra subscribers in Antigravity, the Gemini app and AI Mode in Search. The announcement does not specify a separate price for access through those subscriber products.