Gemma 4 31B

Gemma 4 31B is Google's open model, released 2026-04-02 under Apache 2.0, a standard permissive licence. OpenRouter lists $0.2 per million input tokens and $0.4 output, with a 262,144-token context and image, text and video input. The 26B A4B sibling's name indicates 4 billion active parameters per token, which is aimed at faster serving. We would suggest it for teams that want an open, multimodal model from a large vendor with clear licence terms for local or private use. The limit is that Gemma is smaller than the closed Gemini line, so hard problems may need a stronger model. Our Gemma page covers the E2B to 31B sizes, hardware and the scores Google published. Prices and specifications come from the OpenRouter listing we checked on 31 October 2026, and we have not tested the model ourselves.