Google: Gemma 4 31B (free)

openrouter.ai

Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...

Description

Specs

Type	Model
Section	Models & platforms / text+image+video->text
Pricing	free
Platform	Self-hosted
Systems	api, python, self-hosted
Hosting	cloud
Site language	en
Vendor	Google
Launched	2026-04-02

Can do

text image video

Platforms

api python self-hosted

Found in sources

Similar in «Models & platforms»

Cost-efficient open-source MoE model rivaling GPT-4o in reasoning and math tasks

Open reasoning and coding models on par with the top closed ones.

European open and commercial models, Le Chat and Codestral.

Research faster with Sider AI—summarize web pages, take notes, generate reports, and chat with leading AI models. Powered by ChatGPT, Gemini, Claude, Grok, Deepseek, etc. Try the best AI sidebar now!

The hub of models and datasets: millions of weights, Spaces, Transformers.

Передовая мультимодальная AI-платформа от ByteDance с современными языковыми, визуальными и речевыми моделями, а также интегрированными возможностями рассуждения и поиска.