88Z AI, 19 November — Google has released Gemini 4 Neon, positioned between the flagship Carbon and efficient Hydrogen models. It offers triple the speed of Carbon at 60% of the cost, supports a one-million-token context window and four thinking depth levels, and excels in various benchmarks. Meanwhile, the world model Genie 4 is now available on Project Genie, featuring real-time natural language environment editing during browsing, native audio support, and an Agent for generating and remixing virtual worlds.
At a glance
- Gemini 4 Neon is three times faster than Carbon at lower cost, supporting a million-token context and four thinking depths.
- Genie 4 launches on Project Genie with natural language environment editing, native audio, and virtual world generation.

The update
Google has officially launched its latest flagship model, Gemini 4 Neon, positioned between the top-tier Carbon and the efficient Hydrogen models. This new release balances performance and cost, offering speeds approximately three times faster than Carbon while costing only 60% as much. Technically, Gemini 4 Neon supports a context window of up to one million tokens and provides four adjustable levels of thinking depth. It has achieved leading rankings in key benchmarks including agent collaboration, programming, knowledge work, and cybersecurity. At the same time, Google's world model, Genie 4, is now available on the Project Genie platform. The new version adds the ability to modify virtual environments in real-time using natural language commands during browsing and introduces native audio support. Coupled with the Project Genie Agent, users can more easily generate content and creatively remix or reconstruct virtual worlds.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- 机器之心Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.