Oct 28, 2026
Google's Gemini 4 Carbon evaluated on the Vals Index
We evaluated Google’s new Gemini 4 Carbon on Vals Index v2.2 and its components.
-
#2 of 29 on the Vals Index (70.36%) at $10.51 per test, behind Claude Sonnet 5.5 (70.85%) and ahead of Claude Opus 5.5 (70.18%).
-
Leads on research and finance: #1 of 87 on Legal Research Bench (55.77%), #2 of 35 on EMB v1.1 (76.47%), #3 of 78 on Tax Agent Bench (48.86%) and #5 of 33 on Finance Agent v2.1 (67.16%).
-
Strong on coding: #4 of 57 on Terminal-Bench 4.0 (63.13%) and 89.06% on Vibe Code Bench (#8 of 121). Weakest on HLAB-Vals-1.1 (17.50%, tied #6 of 30).
Gemini 4 Carbon is priced at $4.00 per million input tokens and $20.00 per million output tokens. The model has a 1M-token context window. Evaluations were run with reasoning effort set to “high”.
Congrats to the Google team on the release!