Gemini 4 Hydrogen adds four reasoning levels and model routing
Google’s Gemini 4 Hydrogen targets everyday coding and agent execution, with four reasoning levels from low to xhigh. API pricing is $1.50 per million input tokens and $7.50 per million output tokens across its million-token context window. Google also introduces Antigravity Orchestrator, which switches models within a session instead of assigning every stage to the same model.
Google · Nov 12, 2026
Why it matters
An agent’s planning, implementation and checking stages can have different cost and reasoning requirements.
More context
Available through Google AI Studio, Gemini API and other Google products; Antigravity access is tied to eligible paid subscriptions. Try representative tasks at a fixed reasoning level, then compare end-to-end completion cost. The launch page was updated November 13.
Keep in mind
Released hosted model and vendor-described orchestration. Published speed and quality claims depend on Google’s evaluations; a large context window does not establish reliable use of every token or free API usage.
Original read