The Gemini 4 Neon model supports text, images, audio and video, with a context window of up to one million tokens.
Photo Credit: Google
Gemini 4 Neon can handle longer workflows
Google has unveiled Gemini 4 Neon, its latest artificial intelligence (AI) model, built to tackle demanding coding, research and business tasks. It joins the Gemini 4 lineup between Gemini 4 Carbon, which is designed for more complex workloads, and Gemini 4 Hydrogen, aimed at handling everyday requests quickly. With Neon, the Mountain View-based tech giant is said to be striking a balance between speed and performance while making complex tasks less expensive to run.
Google said in a blog post that Gemini 4 Neon is designed for more demanding tasks, including software development, data analysis, and scientific research. It can handle longer workflows, work with external tools and coordinate smaller AI agents to complete different parts of a task. This sets it apart from Gemini 4 Hydrogen, which is geared towards quicker, everyday requests, while placing it below Gemini 4 Carbon in Google's model lineup.
The Gemini 4 Neon model supports text, images, audio and video, with a context window of up to one million tokens. This lets users work with different types of content in a single session, including PDFs, spreadsheets, charts, and audio or video recordings. Google says Neon can use this information to generate structured reports and support research and business work.
For software developers, Neon can assist with longer coding sessions, from debugging to making changes across multiple files in a project. It can also track progress during these sessions and make targeted edits without repeatedly reading the same files. Its ability to use external tools and coordinate AI agents is intended to help with multi-step tasks.
Google has also shared details on Gemini 4 Neon's performance across independent benchmarks. The model scored 61 on the Artificial Analysis Intelligence Index at its highest thinking level, placing third overall. It ranked second on the Vals Index with a score of 70.9 percent. Google also claims that the model topped several coding and automation tests, including Zapier's AutomationBench.
Gemini 4 Neon users can choose from four thinking levels to control how much reasoning the model applies to a request. Google says the default High setting costs 32 percent less per task than Gemini 4 Carbon, while the higher xhigh setting is meant for more demanding engineering, analytical and scientific work.
The Neon model also has applications in cybersecurity, where it can help identify software vulnerabilities, check whether they can be exploited and suggest fixes. Google is separately preparing Gemini 4 Neon for cyber, a version aimed at cybersecurity professionals. It will be made available to eligible defenders through the company's Fairwind Program.
Gemini 4 Neon is available through the Gemini API in Google AI Studio and Android Studio, as well as Gemini for business. Developers can access Gemini 4 Neon through the paid API, while Google AI Pro and Ultra subscribers can use it in Google Antigravity and the Gemini app, depending on their plan.
For API users, Gemini 4 Neon costs $3 (roughly Rs. 280) per million input tokens and $15 (roughly Rs. 1,400) per million output tokens. Cached input tokens are priced at $0.15 (roughly Rs. 14) per million, a 95 percent reduction compared with standard input pricing.
Get your daily dose of tech news, reviews, and insights, in under 80 characters on Gadgets 360 Turbo. Connect with fellow tech lovers on our Forum. Follow us on X, Facebook, WhatsApp, Threads and Google News for instant updates. Catch all the action on our YouTube channel.
Apple Watch Series 12
Starts from ₹56,900
Samsung Galaxy Watch Ultra 2
Starts from ₹58,949
Samsung Galaxy Watch 9 (44mm, LTE)
Starts from ₹41,999
Samsung Galaxy Watch 9 (40mm, LTE)
Starts from ₹38,999
Samsung Galaxy Watch 9 (40mm)
Starts from ₹36,999
Samsung Galaxy Watch 9 (44mm)
Starts from ₹40,999
Sony WH-CH730N, WH-CH530 Wireless Headphones Launched in India: Price, Specifications
Pay for Siri? iOS 27.3 Could Cap AI Usage Unless You Upgrade to iCloud+
GTA 6 Download Size on PS5 and Xbox Series X/S: How Much Space Do You Need?