Gemma 3 4B

Google DeepMind · google/gemma-3-4b-it

ProviderGoogle DeepMind
Familygemma
Typevlm
Statusactive
Released2025-03-13
Updated2025-03-13

Reported benchmark scores

These scores carry one collection date for the whole card (2026-04) and a source list rather than a source per score, so they cannot be attributed individually. They are shown as reported and are not verified evidence. Benchmark pages carry the reviewed, dated evidence where it exists.
BenchmarkCatalogue standingScore
AI2D (AI2 Diagrams)unassessed74.8
Arena Elo — Codingunassessed1190.0
Arena Elo — Hard Promptsunassessed1283.4
Arena Elo — Mathunassessed1170.0
Arena Elo — Overall (Text)unassessed1180.0
Arena Elo — Style Controlunassessed1302.8
ChartQAunassessed68.8
DocVQAunassessed75.8
GPQA Diamondunassessed33.2
GSM8Kunassessed89.2
HumanEvalunassessed58.5
IFEvalunassessed62.0
LiveCodeBenchunassessed12.6
MATH-500unassessed48.5
MathVistaunassessed50.0
MMLU-Prounassessed43.1
MMMUunassessed48.8

Data

This card as JSON · Edit on GitHub