Available Models
Open AI Catalog in Bunyan to see the model configurations available to you. The list depends on your workspace, access, and the configurations registered there.
Find a Configuration
- Check your selected organization and project.
- Open AI Catalog and search for a model.
- Narrow the list with the provider, source, and type filters. Sources include Bunyan, Shared, and Private.
- Review the configuration's model type and settings before choosing it for your task.
Model Types
| Type | Use it for |
|---|---|
| Chat | Text conversations and generation. |
| Multimodal | Requests combining supported input types, such as text and images. |
| Embedding | Vector representations for search and retrieval. |
| Text-to-Speech | Generating speech from text. |
| Speech-to-Text | Transcribing audio. |
| Image Generation | Generating images. |
A model type describes the configuration's purpose. Supported inputs, settings, and limits depend on the selected model.
Use an Available Model
For conversations, select an available configuration in the Playground. To call an LLM configuration from your application, follow LLM Endpoints and use its configuration name.
If a configuration is missing, check your workspace and ask its owner or your administrator to review your access.
Deploying a Model
To provision an inference server, open the deployment page and select Choose from Catalog. This is a separate catalog of deployment options. See Deploying AI Models.
To connect an existing model endpoint, see Add Your Own Model.