Overview
- Google launched Gemini 4 Neon on Wednesday, November 18, and began rolling it out through the Gemini API, Google AI Studio, Antigravity and the Gemini app for paid AI Pro/Ultra and business customers.
- Neon’s published API rates are $3 per million input tokens and $15 per million output tokens with cached inputs discounted to $0.15 per million, a pricing structure that affects the cost of every call because developers pay for both input and output tokens.
- Google positions Neon as a mid‑tier Gemini 4 model with a 1,000,000‑token context window and four thinking levels to trade reasoning depth for cost, claiming Neon delivers near‑Carbon reasoning at higher throughput.
- Google published benchmark tables showing strong Neon results — including a 61 score on the Artificial Analysis Intelligence Index and a 1336, #2 placement on Arena.ai’s Agent Arena Index — but broad independent verification of all claims remains incomplete.
- A gated 'Neon for cyber' variant will be provided to vetted security defenders through Google’s Fairwind Program in the coming days, and scheduled December 1 price changes for Argon and Carbon will materially alter Neon’s relative cost advantage for production workloads.