Perceptron Mk1
🧠 AI Modelperceptron
Top-tier vision-language model for video and embodied reasoning.
Perceptron Mk1 (Mark One) is Perceptron's flagship vision-language model designed for video and embodied reasoning. It processes multimodal inputs—text, images, and videos—and generates text outputs with detailed visual understanding. The model boasts a 32,768-token context window, enabling long-form video analysis. Pricing is $0.15 per million input tokens and $1.50 per million output tokens. It supports advanced features like structured outputs, reasoning, frequency/presence penalties, and temperature control. This makes it suitable for complex tasks such as video question answering, robotics planning, and interactive AI systems that require grounding in visual data.
💡Highlights
- ├─32,768 context length
- ├─Video & image understanding
- └─Structured outputs support
🎯For
- ├─AI Researchers
- ├─Computer Vision Engineers
- └─Robotics Developers