Gemini 3 Pro Preview
Google · 2 thinking configurations
Intelligence22.3
Input / 1M$2.00
Output / 1M$12.00
Context1M
Tariff: Google · Source: artificial-analysis · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost2 measured configurations · 0 without both values
Selected: thinking low · Click a point to switch configuration
Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
22.3 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
88.7 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
29.5 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
68.1 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
49.7 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
48.3 %Specifications from the catalog
Inputtext, image, video, audio, pdf
Outputtext
Tool callingYes
Structured outputYes
SpeedNot available
First tokenNot available
Maximum output65.5K
Catalog listing date2025-11-18