Image-Text-to-Text

Vision modality not enabled

#19
by Deeko76 - opened

Pushed the model to a basic Android app and when trying to use an image as input with correct sizing it returns an error stating Vision modality is not enabled?

Google org

Hi @Deeko76 ,

This looks like you haven't enabled vision support for LLM Inference API, you have to set EnableVisionModality configuration option to true within the Graph options.
Here is the official reference: https://ai.google.dev/edge/mediapipe/solutions/genai/llm_inference/android#image-input

Thank you!

Sign up or log in to comment