priorsBuilder

Google · released 2026-02-12

Gemini 3

Measured on 4 benchmarks.

General frontier

Not ranked in this mix — too little measured data for its benchmarks.

Your mix’s benchmarks

0 of 8 measured

This model has no results on your mix’s benchmarks.

All 4 benchmarks

show 4 more

creative

VoxelBench (text)

55 results
  1. #37default of 551422

Leader: GPT-6 Astra · 2732

reasoning

ARC-AGI-1

215 results
  1. #27default of 21596.0%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #30default of 21684.6%

Leader: GPT-6 Astra · 95.0%

reasoning

CritPt

529 results
  1. #41default of 52925.7%

Leader: Muse Wonder Full (Max) · 32.9%