Skip to content

Starts the installed server as a managed child of the engine and returns at once; the row reads "running" once the server answers.

POST
/v1/m/agenttools/model-servers/{kind}/start
curl --request POST \
--url https://example.com/v1/m/agenttools/model-servers/example/start \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{ "model": "example", "tenant_id": "2489E9AD-2EE2-8E00-8EC9-32D5F69181C0" }'

Starts the installed server as a managed child of the engine and returns at once; the row reads “running” once the server answers.

kind
required
string

Path parameter kind.

Media type application/json
object
model
required

Llama-cpp: the absolute path of a .gguf file, or a Hugging Face model as user/repo[:quantization]. vllm: the absolute path of a model directory, or a Hugging Face model as user/repo.

string
<= 1024 characters
tenant_id
required

The organization whose Providers get this server’s endpoint and the key the engine made for it.

string format: uuid
Example generated
{
"model": "example",
"tenant_id": "2489E9AD-2EE2-8E00-8EC9-32D5F69181C0"
}

The server is starting, or already runs this model, on this node

Media type application/json
object
endpoint
string
installed
required
boolean
kind
required
string
Allowed values: llama-cpp vllm
message
string
model
string
models
required
Array<string>
name
required
string
state
required
string
Allowed values: stopped starting running failed
Example
{
"kind": "llama-cpp",
"state": "stopped"
}

Bad request

Media type application/json
object
Example generated
{}

Unauthenticated

Media type application/json
object
Example generated
{}

Forbidden / step-up required

Media type application/json
object
Example generated
{}

Not found

Media type application/json
object
Example generated
{}

Conflict

Media type application/json
object
Example generated
{}

Required evidence or store unavailable

Media type application/json
object
Example generated
{}