Model Serving¶
Model Serving allows you to quickly deploy open-source, self-developed, or fine-tuned large language models as inference services that can be called within the platform.
- Users can call the selected model service to perform tasks such as text generation, image understanding, and image generation.
- Online model trials are supported.
- API calls and service O&M management are supported.
Common Operations¶
- Deploy a new model: See Deploying a New Model.
- Manage deployed services: See Manage Model Services.
- View deployment details (authentication and API examples): See Model Service Details.
- Obtain an API Key: See API Key Management.
- View model invocation instructions: See Model Invocation.