Rankings / Google DeepMind

Gemma 4 31B IT

released 2026-04-02

BenchAtlas Index

as of 2026-09-11

Not enough benchmark coverage for a composite score (needs ≥4 families across ≥3 categories). The per-benchmark evidence below stands on its own — sparse coverage is an evidence gap, not a low score.

Benchmark evidence

7 results

external indices

ECI	minimal effort	142.7 points	independent	Epoch AI Benchmarking Hub
ECI		142.7 points	independent	Epoch AI Benchmarking Hub

factuality

SimpleQA Verified		10.4 %	independent	Epoch AI Benchmarking Hub 2026-08-27
SimpleQA Verified		9.6 %	independent	Epoch AI Benchmarking Hub 2026-04-04

knowledge science

GPQA Diamond

minimal effort

75.8 %

independent

Epoch AI Benchmarking Hub

2026-08-06

professional

CaseLaw

high effort

52.6 %

independent

Vals AI

reasoning math

OTIS Mock AIME 2024–2025

minimal effort

73.3 %

independent

Epoch AI Benchmarking Hub

2026-08-06