Probe embedding models¶
Requires regular access to the current workspace.
POST https://api.moi.matrixorigin.cn/v5/embedding/probe-models
Before you call this API¶
Set BASE_URL to the product API prefix https://api.moi.matrixorigin.cn/v5; use the corresponding product API URL for a dedicated deployment.
Configure identity, workspace, and model information in the same environment.
Request¶
curl --request POST \
"$BASE_URL/embedding/probe-models" \
--header "X-API-Key: $MOI_KEY" \
--header "X-Workspace-ID: $WORKSPACE_ID" \
--header "Content-Type: application/json" \
--data-raw '{"address":"https://provider.example.invalid/v1","api_key":"provider-key-example"}'
Parameter |
Location |
Type |
Required |
Description |
|---|---|---|---|---|
|
Headers |
string |
Yes |
Workspace ID in the same environment. |
|
Body |
string |
Yes |
Provider API base URL; the product trims whitespace and appends /models. |
|
Body |
string |
No |
Provider model-list credentials; optional for services requiring no key, and separate from MOI_KEY. |
Requires regular access to the current workspace. Sends GET /models to the provider, with Bearer when a key is supplied; creates no backend. Name patterns distinguish LLM/Embedding without proving actual capabilities. Duplicate candidates are not necessarily merged. Non-200 responses, invalid responses, and size-limit violations fail.
Successful response¶
{
"code": "OK",
"msg": "OK",
"data": {
"models": [
"bge-m3"
]
}
}
Field |
Type |
Description |
|---|---|---|
|
string |
OK on success. |
|
object |
Error responses¶
States and errors distinguishes string and numeric protocols. Confirm probe candidates, persistence, and actual inference separately.