Gemini 4 Hydrogen beats Argon at 51 vs. 50, priced at $1.50

Gemini 4 Hydrogen arrives as a complement to Google's frontier models, Gemini 4 Carbon and Gemini 4 Argon, which the company says excel at deep, long-horizon reasoning. Hydrogen is positioned for rapid execution across agentic workflows, and Google is making it available across all tiers of its developer, enterprise, and consumer surfaces simultaneously.

The model's benchmark positioning is notable: it surpasses Argon on the Artificial Analysis Intelligence Index at its default high thinking level, 51 versus 50, and reaches 54 at its xhigh setting. That suggests Google is aiming to deliver top-tier intelligence without the latency or cost typically associated with frontier reasoning models. The flat pricing of $1.50 per million input tokens across a 1 million-token context window reinforces that positioning.

For teams migrating from Gemini 3.8 Flash, Google presents Hydrogen as a major standalone upgrade, claiming it finishes tasks 2.5 times faster at its high default setting and more than triples Terminal-Bench 4.1 scores. The company also emphasizes safety improvements, including reduced overreach in agentic tasks by up to 81% and a 96.7% honesty rate on its internal agent honesty evaluation, framing Hydrogen as designed for dependable execution at scale.