Skip to content

Gemma 4 12B (Non-reasoning)

Available

Google · 2026-06-03 · 32,000 tokens

An AI model from Google, suited to a broad range of AI workloads.

Supported modalities:textcode

Quick Overview

Text Generation1/10
Code Generation6/10
Reasoning6/10
Multimodal1/10

Benchmark Results

Scores from leading benchmark suites.

artificial analysis intelligence13.2

Performance Metrics

Latency and throughput performance.

P50 Latency
138.236tokens/sec

Dive Deeper