Moondream 3.1 is a vision language model featuring a mixture-of-experts architecture with 9 billion total parameters and 2 billion active parameters. It aims to deliver state-of-the-art visual reasoning and detection capabilities while remaining fast and cost-effective for deployment.
- The model utilizes a mixture-of-experts architecture (9B total, 2B active).
- It provides native skills including query, detect, point, and caption.
- All skills return structured output.