Google: Gemma 4 31B
google/gemma-4-31b-it
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.
- Context window: 262,144 tokens
- Input: image, text, video
- Output: text
- Pricing: $0.09/M input tokens, $0.34/M output tokens
RuntimeWire coverage
- Tomas Koutsky explains how Meta fits Muse Glimmer into 24 GB
- Jeff Morgan's Ollama raises $65M as local AI becomes a cloud business
- Pearl's AI mining pitch faces a 112 MW usefulness test
- Nemotron-3 Ultra crushes Gemma-4 31B by 6 points
- The Week Open Weights Went Multimodal
- Ollama QAT Weights Put Claimed Gemma 4 31B Scores Near Claude Opus 4
- NVIDIA's new Nemotron model takes the top US open-weight slot, Artificial Analysis says
View on OpenRouter. Model data sourced from OpenRouter.