Nov 12, 2026
Google's Gemini 4 Hydrogen is #1 on Tax Agent Bench
We evaluated Google’s new Gemini 4 Hydrogen on Vals Index v2.2, its components and the Vals Finance Index and Vals Legal Index.
-
#1 of 85 on Tax Agent Bench (52.30%), ahead of Gemini 4 Carbon (51.78%) and Claude Fable 5.1 (50.75%).
-
#2 of 40 on the Vals Finance Index (66.14%), behind Gemini 4 Carbon (67.05%).
-
#6 of 37 on the Vals Index (68.49%) at $9.84 per test, behind Claude Fable 5.2 (71.31%).
-
Top 6 on Legal Research Bench (53.85%, #6 of 94) and EMB v1.1 (75.80%, #6 of 42).
-
Mid-pack on coding: #11 of 64 on Terminal-Bench 4.0 (54.04%) and #13 of 128 on Vibe Code Bench (87.78%).
Gemini 4 Hydrogen is a reasoning model with a 1M-token context window, priced at $1.50 per million input tokens and $7.50 per million output tokens. We evaluated it at xHigh thinking effort.
Congrats to the Google team on the release!