Gemma 3 12B

Google's 12-billion-parameter open-weight text model with a 131K token context window, released in 2025 between the Gemma 3 4B and Gemma 3 27B.

Strengths

Offers a middle ground between lightweight and larger open-weight models, with expanded context compared to earlier Gemma releases.

Best for

Deployments where you need more capacity than Gemma 3 4B but want to avoid the overhead of running Gemma 3 27B or larger models.

Limitations

Smaller than Gemma 4 31B released later in 2026, and has a narrower context window (131K vs 262K) than its successor.