Google Introduces Gemini 4 Neon for Advanced AI Tasks
Google has introduced Gemini 4 Neon, a new artificial intelligence model designed to handle complex coding, research, business, and scientific tasks while offering faster performance and lower operating costs than larger AI models. The company states that Neon combines advanced reasoning capabilities with efficient task execution, making it suitable for software development, enterprise workflows, and long-running AI agents. According to early evaluations by Glean, Gemini 4 Neon matched or exceeded Gemini 4 Carbon in overall quality while operating three times faster and at 60% of the cost. The model is rolling out through the Gemini app, Google AI Studio, Gemini API, Google Antigravity, and Google's business AI services.
Positioned within the Gemini 4 lineup between Carbon and Hydrogen, Neon balances advanced reasoning, speed, and operating costs. Carbon is designed for the most demanding long-running projects, whereas Hydrogen focuses on everyday tasks. The new release aims to make high-level artificial intelligence reasoning accessible across multiple consumer, developer, and enterprise platforms.
Performance Benchmarks and Evaluation Results
Google reports that Gemini 4 Neon ranks third overall on the Artificial Analysis Intelligence Index, scoring 61 at its highest thinking level. The model also ranks second on the Vals Index, with a score of 70.9%, and second on Arena's Agent Arena Index, scoring 1,336 across more than 4,200 real-world AI agent sessions. On the new Agent Arena steerability evaluation, Neon ranks second for adapting to user corrections during ongoing tasks. Google also reports strong performance in financial analysis, accounting, spreadsheet modeling, legal research, and tax-related tasks.
The model ranks first on the Legal Research Bench and Tax Agent Bench, while achieving 73.5% on Chartography and 75.8% on Agents' Last Exam v1.1. It also leads Zapier's AutomationBench with 56.3% and Andon Labs' Vending-Bench 2, where it earned $16,843 in a simulated business environment. In scientific reasoning, Google reports scores of 70.5% on Terminal-Bench Science, 84% on RiemannBench, and 71.3% on HLE-Diamond. These results come from a combination of independent evaluations and benchmarks reported by Google.
Coding Improvements and Agent Coordination
Gemini 4 Neon is designed for complex software development, including debugging, refactoring, working across large codebases, and coordinating multiple AI agents. The model ranks first on the Artificial Analysis Coding Agent Index with a score of 58.6. It also achieves 65.2% on Terminal-Bench 4.1 and 46.2% on FrontierCode 2.0 at its default high thinking level. On SWE-sweep, Neon scores 12.8% using the mini-SWE harness, increasing to 39% when operating through Google Antigravity.
Google states that the model performs better during long-running tasks by reducing unnecessary file reads, maintaining context, and sharing progress updates. In Antigravity, Neon can also coordinate multiple Gemini 4 Hydrogen subagents to complete larger projects. Google demonstrated these capabilities through several projects, including a photorealistic 3D laboratory built using Blender, a WebGL2 skiing simulator based on real Mont Blanc terrain, and an interactive recreation of Alan Turing's 1936 study. Another demonstration showed Neon converting a single image into a Google Sheet containing 100,000 cells and more than 500,000 formulas, without embedded images or Apps Script.
Cybersecurity Capabilities and Fairwind Program
Google has also improved Neon's ability to identify and fix software security vulnerabilities. The model achieved 76% on the hardest AutoBaxBench security challenges and 51.26% on WebExploitBench v1.2, with Google reporting lower costs than competing large models. It also scored 90.5% on SEC-bench Pro for identifying vulnerabilities and 70.7% on Collinear's CWE-bench v2 for patching them. On ExploitGym, Neon demonstrated competitive performance in validating whether vulnerabilities could be exploited while using half the tokens required by Gemini 4 Carbon.
Google states that these improvements could help security teams identify, assess, and repair vulnerabilities more efficiently. A specialized version, Gemini 4 Neon for cyber, will become available to eligible security professionals through Google's Fairwind Program over the coming days. Eligible cybersecurity professionals will begin receiving access to Gemini 4 Neon for cyber as part of Google's broader strategy to expand specialized AI deployment.
Pricing Structure and Four Thinking Levels
Google has priced Gemini 4 Neon at $3 per million input tokens and $15 per million output tokens. Cached input tokens cost $0.15 per million, representing a 95% discount compared with standard input pricing. The model also supports a 1 million-token context window, allowing it to process large amounts of information across documents, spreadsheets, charts, images, audio, and video.
Neon offers four thinking levels, allowing developers to adjust reasoning effort based on task complexity and cost requirements. According to Google, its default high thinking level costs 32% less per task than Carbon on the Artificial Analysis Intelligence Index. The company also claims Neon can cost approximately 50% to 80% less per task than leading competing models on certain evaluations. At its highest thinking level, Neon delivers its strongest reasoning results while reportedly costing at least 36% less per task than competing models at their highest settings.
Reliability, Safety, and User Capabilities
Google reports that Gemini 4 Neon has achieved its lowest fabrication rate yet in internal Google Antigravity testing. The model also showed improvements in instruction following, resistance to malicious prompts, and staying within user-defined boundaries. In Google's internal evaluations, Neon achieved a 99.2% pass rate on agent honesty tests and a 0.09% misaligned-action rate. It recorded no cheating on Terminal-Bench 4.0 and took shortcuts in 2.45% of tasks in Google's coding evaluation.
In the Gemini app, Neon is intended to handle complex research, document analysis, writing, and multi-step tasks. Users can upload spreadsheets, PDFs, images, and other files, then ask the model to extract information, generate reports, or perform detailed analysis. The model can also use information from connected Google services such as Gmail and Docs. Google notes that Neon can help with practical tasks such as reviewing billing statements, identifying unusual charges, comparing vendor prices, and analyzing documents for refund requests.
Availability Across Google Ecosystem
Gemini 4 Neon is rolling out across Google's AI products. Developers can access the model through the paid Gemini API in Google AI Studio and Android Studio. Google AI Pro and Ultra subscribers can use it in Google Antigravity, while Pro and Ultra subscribers can select Neon from the model menu in the Gemini app. Enterprise customers can access the model through Gemini for business using the API or model selector. Google states that the release is aimed at making advanced AI reasoning and long-running task execution faster and more affordable for developers, businesses, and consumers.