Description
When updating an endpoint, all parameters must be provided in full.
Bearer authentication format, for example: Bearer {{API Key}}.
Request Body
The ID of the endpoint to update.
Cluster information. Required when mounting cloud storage, and must match the cluster ID where the cloud storage resides. String, length limit: 0-255 characters.
Worker configuration. The valid range can be dynamically obtained via the parameter range API. Minimum number of workers.
Maximum number of workers.
Idle timeout (seconds). Unit: seconds.
Number of GPUs per worker.
HTTP ports. Only one is supported. Supported port range: 1-65535, except for 2222, 2223, and 2224 which are reserved for internal use.
Auto-scaling policy. The valid range can be dynamically obtained via the parameter range API. Policy type. Options:
queue
: Queue latency policy, adjusts the number of workers based on the waiting time of requests in the queue.
concurrency
: Queue request policy, automatically adjusts the number of workers based on the number of requests in the queue.
The meaning of value depends on the type:
If type = queue, value is the queue waiting time in seconds.
If type = concurrency, value is the maximum number of requests in the queue.
Image information. Image URL. String, length limit: 0-511 characters.
Private image credential ID (not required for public images or platform user images). String, length limit: 0-255 characters.
Container startup command. String, length limit: 0-2047 characters.
Storage information (GB). Storage type. Options:
local
: Local storage.
network
: Network storage.
Local storage size, currently set to a fixed value of 30. Not required for network storage.
Network storage ID. Not required for local storage.
Mount path. String, length limit: 0-255 characters.
Environment variables. Environment variable name.
Environment variable value.
Health check endpoint path.