Google has released Gemini 4 Hydrogen, its newest AI model, making it available globally across Google Search, the Gemini app, Google AI Studio, the Gemini API, Gemini for business, and Google Antigravity. This update marks the largest model upgrade to Google Search to date, replacing the previous Gemini 3.8 Flash model.
The new model is now the default for difficult queries in AI Overviews and is designed for fast execution across various agentic workflows, where AI carries out multiple steps independently.
Gemini 4 Hydrogen offers significant advancements in speed and intelligence compared to its predecessor. It completes tasks 2.5 times faster at its high default setting and reduces fabricated claims by more than 5 times. The model also triples Terminal-Bench 4.1 scores and shows improved adaptability when tool calls fail or searches return empty.
Hydrogen features native 1 million-token input and up to 1 million output tokens. It surpasses Gemini 4 Argon on the Artificial Analysis Intelligence Index at its default high thinking level (51 vs. 50), reaching 54 at its highest 'xhigh' setting.
In Google Search, Gemini 4 Hydrogen provides a better understanding of user intent, allowing it to handle more nuanced and complicated questions, including those with multiple parts, typos, or highly specific requests. It more precisely follows user instructions and delivers relevant responses.
The model also improves query fan-out, connecting users to a broader range of helpful links by conducting simultaneous queries across relevant subtopics. Responses now include more visuals, such as images, comparison tables, graphs, and maps.
Alongside Gemini 4 Hydrogen, Google introduced an Orchestrator system within its Antigravity development environment. This system dynamically switches between Gemini 4 Carbon and Hydrogen models during coding tasks. Carbon handles demanding reasoning like architecture and debugging, while Hydrogen manages code changes and testing.
This combined approach achieves nearly the same results as using Carbon alone, but with 24% less quota usage and 19% faster task completion, representing a significant advancement for agentic coding.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's briefNew every morning, and the back catalogue is archived by date.
Google has released its new Gemini 4 Hydrogen model, which provides more helpful answers to complex questions by breaking them down and addressing each part. This model is now available across Google Search, the Gemini app, Google AI Studio, and other Google AI services, becoming the default for difficult queries in AI Overviews. It also offers a wider range of links, more visuals, and improved language understanding.
Google launched Gemini 4 Hydrogen and a new Orchestrator system that dynamically switches between Gemini 4 Carbon and Hydrogen models during coding tasks. This approach achieves nearly the same results as using Carbon alone, but with 24% less quota usage and 19% faster task completion. The Orchestrator feature in Google's Antigravity environment assigns demanding reasoning to Carbon and coding/testing to Hydrogen.
Google Search has integrated the new Gemini 4 Hydrogen model to better understand complex queries, including multi-part questions, specific requests, and typos. This update aims to provide more accurate and visually rich answers, and faster processing compared to previous models.
Google has released Gemini 4 Hydrogen, which replaces the Gemini 3.8 Flash model across its products. Hydrogen offers significant speed improvements, reduces fabricated claims, and shows gains in code generation compared to its predecessor.
Google released Gemini 4 Hydrogen, a new AI model that replaces the "Flash" branding and offers significant advancements in speed and intelligence. This model features native 1 million-token input and output, completing tasks 2.5 times faster and tripling Terminal-Bench 4.1 scores compared to its predecessor. Gemini 4 Hydrogen also reduces fabricated claims by over 5x and shows improved adaptability when tool calls fail.
Google has launched Gemini 4 Hydrogen, an AI model designed for rapid execution in agentic workflows and everyday coding tasks. It is available across Google AI Studio, the Gemini API, Google Search, and for paid subscribers, complementing the more complex Gemini 4 Carbon and Argon models. Hydrogen offers a flat pricing structure and integrates with a new Orchestrator tool for multi-model efficiency.
Google Search has integrated Gemini 4 Hydrogen, its largest model upgrade to date, globally for all users. This update provides improved intent understanding, broader query fan-out, more visual responses, and richer language understanding in AI Mode.