Editorial illustration for Google Ships Gemini 4 Neon, Its Fourth New Gemini Model in a Month
Google Launches Gemini 4 Neon, Fourth Model This Month
Google put out Gemini 4 Neon on its API today, charging $3 per million input tokens and $15 per million output tokens. That's pricier than Carbon and Argon, which launched at an introductory $2/$10 rate before doubling to $4/$20 on December 1, and double what Hydrogen costs at $1.50/$7.50. Neon handles a roughly one-million-token context window and can output up to a million tokens itself, working across text, images, video, audio, and PDFs. It also supports Computer Use, code execution, file search, and grounding through Google Search and Maps, though image and audio generation are off the table.
Google's pitch centers on "frontier-level intelligence with fast end-to-end task completion," aimed at software engineering, knowledge work, and science tasks. Benchmark numbers haven't landed yet but are expected within hours, and full technical specs are already posted in Google's documentation. What's notable here isn't just the feature set.
Neon marks the fourth distinct Gemini 4 model Google has shipped in four weeks, following Hydrogen, Carbon, and Argon in quick succession. That pace raises the question of where Neon actually sits in Google's lineup compared to the rivals it's chasing.
Google released Gemini 4 Neon, promising "frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work, and science workflows." The model is live today through the API at $3 per million input tokens and $15 per million output tokens, above Carbon and Argon's introductory rate of $2/$10 (doubling to $4/$20 on December 1) and twice the price of Hydrogen at $1.50/$7.50.
Why this matters
Four model names in a month is a lot to track, and Google hasn't made the case for why Neon needs to exist alongside Carbon, Argon, and Hydrogen rather than replacing one of them. The pricing alone tells a story: at $3/$15, Neon costs twice what Hydrogen charges and 50% more than Carbon and Argon's introductory rate, with that gap set to narrow once those two jump to $4/$20 on December 1. For teams building on Gemini, this is now a real procurement question, not just a model-selection one.
Which tier do you build against when the cheapest option today might cost double in two weeks, and the priciest option today might look cheap by comparison? We'd treat "frontier-level intelligence" claims with the usual caution until independent benchmarks surface, but the bigger signal here is velocity: Google is iterating faster than most teams can evaluate, let alone migrate. If you're building production systems on Gemini, the smart move is pinning versions and watching the pricing calendar closely, because the naming scheme alone won't tell you what you're actually paying for next month.
Common Questions Answered
What are the pricing differences between Gemini 4 Neon and other recent Gemini models?
Gemini 4 Neon costs $3 per million input tokens and $15 per million output tokens, making it twice as expensive as Hydrogen ($1.50/$7.50) and 50% more expensive than Carbon and Argon's introductory pricing of $2/$10. Carbon and Argon's prices are set to double to $4/$20 on December 1, which will narrow the gap with Neon but still keep Neon as the premium option among the four models.
What capabilities does Gemini 4 Neon support across different media types?
Gemini 4 Neon handles a roughly one-million-token context window and can process and output content across text, images, video, audio, and PDFs. The model also supports Computer Use and code execution capabilities, enabling it to handle complex software engineering, knowledge work, and science workflows.
Why is Google releasing four different Gemini models within a single month?
Google has not clearly communicated the distinct use cases or advantages that justify having Gemini 4 Neon, Carbon, Argon, and Hydrogen as separate offerings rather than consolidating them. The pricing structure suggests different performance tiers, but the rapid succession of releases creates confusion for teams trying to determine which model best fits their needs and budget.
What is the context window size and output capacity of Gemini 4 Neon?
Gemini 4 Neon handles a roughly one-million-token context window and can output up to a million tokens itself. This large context window enables the model to process and generate substantial amounts of information for complex tasks.
Further Reading
- Introducing Gemini 4 Neon - Google Blog
- Gemini 4 Carbon is now available - Google Blog
- Gemini 4 Carbon: our next frontier model - Google Blog
- Google launches Gemini 4 Neon with Carbon quality at Flash speeds, live for AI Pro - 9to5Google
- What's new in Gemini 4 Neon - Google AI Studio