Google rolls out Gemini 4 Neon, with "fast end-to-end task completion" across coding, knowledge work, and science, in the API, Antigravity, and the Gemini app
First reported by Blog.google ·
The price for using the Gemini API for model output tokens drops to $15 per million.
Google has released Gemini 4 Neon, a new AI model emphasizing "fast end-to-end task completion" across software engineering, knowledge work, and scientific tasks. The model demonstrates competitive performance, ranking highly on benchmarks like the Artificial Analysis Intelligence Index and Vals Index, and also shows strong steerability and reduced fabrication rates in internal testing. Gemini 4 Neon is integrated into the Gemini API, Google AI Studio, Gemini for business, and the Gemini app for paid subscribers. It is designed to synthesize information from diverse inputs including spreadsheets, PDFs, audio, and video within a 1 million-token context window. The model is engineered for long-running agent workflows, supporting complex, tool-heavy tasks and aims to provide efficiency and control for production-grade workflows with tiered thinking levels.
Gemini 4 Neon positions itself as a highly capable and efficient AI model, excelling in complex tasks across various domains from coding to scientific research. Its notable performance on benchmarks, particularly in steerability and multi-step instruction following, suggests it is designed for robust agentic workflows. The model's ability to handle diverse data formats and a large context window means it can tackle more comprehensive and integrated tasks. Google's continued focus on production-readiness, efficiency, and cost-effectiveness with Gemini 4 Neon indicates a strategic push to make advanced AI more accessible and practical for enterprise and developer use cases.
The introduction of Gemini 4 Neon signifies a market trend towards more integrated and performant AI agents that can handle a wider array of complex tasks with greater efficiency. By offering frontier-level intelligence at a competitive price point, Google is likely aiming to democratize access to advanced AI capabilities, impacting developers and businesses that rely on AI for complex problem-solving. The emphasis on "fast end-to-end task completion" and "production-ready code" suggests a move beyond basic AI functions towards more sophisticated applications that can directly contribute to operational workflows and product development.
AI-written summary. May contain errors.