Quickstart

Start calling this model with endpoint-specific examples.

Benchmarks

Headline benchmark standings and comparison context.

About

Key dates, capabilities, and model metadata.

Quickstart

Start calling this model with endpoint-specific examples.

Benchmarks

Headline benchmark standings and comparison context.

About

Key dates, capabilities, and model metadata.

Google: Gemma 4 12B

Chat Compare

Gemma 4 12B is a dense, encoder-free multimodal model from Google DeepMind built for local AI workloads. It accepts text, image, audio, and video inputs and produces text outputs.

Overview Playground Providers Pricing Performance Apps Activity Quickstart Benchmarks Family Timeline

No family information yet

We haven't linked this model to a wider family. Check back soon as we expand our coverage.