Local inference
The model runs on the appliance instead of an external API.
On-premise model inference
GeoAI
An on-premise model service with chat, deep thinking and image understanding.
GeoAI runs a model service inside the customer network. It is used from the console for chat, with streaming responses, deep thinking and image understanding.
Because inference runs locally, prompts and documents do not need to leave the environment.
Capabilities
Streaming chat
Responses stream into the console as they are generated.
Deep thinking
A reasoning mode for longer questions and analysis.
Image understanding
Accepts images alongside text for visual questions.
Put the product into your scenario
We can arrange a demo around your data and deployment conditions.