This guide explains how to enable and configure automatic AI powered captions, which is part of the Helper feature of Discourse AI plugin.
Required user level: Administrator
Auto caption allows users to harness AI to automatically caption any uploaded image, providing detailed descriptions.
Features
- Manual
Caption with AIoption - Automatically caption with AI
- Alerting users of enabling Auto caption for all uploaded images
- Editable caption always including “(Captioned by AI)” in copy
Enabling Auto caption
Prerequisites
You must configure at least one vision-powered Large Language Model (LLM) from a provider.
To get started you can configure them through the Discourse AI - Large Language Model (LLM) settings page.
- OpenAI
- Anthropic
- Azure OpenAI
- AWS Bedrock with Anthropic access
- HuggingFace Endpoints with Llama2-like model
- Self-Hosting an OpenSource LLM
- Google Gemini
- Cohere
Configuration
- Go to Admin → Plugins
- Search for
discourse-aiand ensure it’s enabled - Ensure the Helper feature is enabled by checking
composer ai helper enabled - In
ai helper enabled featuresensureimage_captionis an enabled feature of Helper - Select the LLM at
ai helper image caption model - We recommend setting
ai auto image caption allowed groupsto specific user groups so they can use Auto caption
Additionally, users can also set their individual preferences to enable/disable Auto caption by navigating to their own user profile → preferences → AI
Technical FAQs
Why is the caption description not matching the image?
- This depends on the vision-powered LLM and the image itself. More powerful LLMs might be able to better describe the picture but that is dependendent on the model. Picture fidelity and the subject of the picture are also important factors to consider in caption output.
Last edited by @Saif 2024-11-14T15:50:24Z
Check document
Perform check on document: