List AI Studio service slots¶
View the model allowlist, default model, and selection mode for each feature. Unconfigured slots are also returned; configured does not establish model health.
GET https://api.moi.matrixorigin.cn/v5/admin/aistudio/service-slots
Before you call this API¶
Prepare a Genesis administrator Bearer token valid for the current environment. An ordinary personal access token, X-API-Key, or Workspace Owner status is insufficient. Session authentication also requires the corresponding CSRF protection. See AI Studio. These are account-level administration operations and do not require a Workspace ID.
Request parameters¶
curl --get "https://api.moi.matrixorigin.cn/v5/admin/aistudio/service-slots" \
-H "Authorization: Bearer $GENESIS_ADMIN_TOKEN"
Parameter |
Type |
Required |
Description |
|---|---|---|---|
|
string |
Yes for the Bearer example |
Bearer followed by the current environment’s administrator token. Do not also send X-API-Key. |
Successful response¶
{
"slots": [
{
"slot_id": "parse",
"name": "Document parse",
"frontend_behavior": "hidden",
"allowed_model_types": [
"chat_multimodal",
"ocr"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "audio_parse",
"name": "Audio parse",
"frontend_behavior": "hidden",
"allowed_model_types": [
"asr"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "extract",
"name": "Information extract",
"frontend_behavior": "hidden",
"allowed_model_types": [
"chat_text",
"chat_multimodal"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "img_parse",
"name": "Image parse",
"frontend_behavior": "hidden",
"allowed_model_types": [
"chat_multimodal",
"ocr"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "ai_infer",
"name": "AI inference",
"frontend_behavior": "selectable",
"allowed_model_types": [
"chat_text",
"chat_multimodal"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "embed_text",
"name": "Text embedding",
"frontend_behavior": "hidden",
"allowed_model_types": [
"embedding_text"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "embed_img",
"name": "Image embedding",
"frontend_behavior": "hidden",
"allowed_model_types": [
"embedding_multimodal",
"embedding_text"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "rerank",
"name": "Rerank",
"frontend_behavior": "hidden",
"allowed_model_types": [
"rerank_text",
"rerank_multimodal"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "workflow",
"name": "Intelligent workflow",
"frontend_behavior": "selectable",
"allowed_model_types": [
"chat_text",
"chat_multimodal"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "nl2sql",
"name": "NL2SQL",
"frontend_behavior": "hidden",
"allowed_model_types": [
"chat_text",
"chat_multimodal"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "knowledge_qa",
"name": "Knowledge Q&A and RAG",
"frontend_behavior": "selectable",
"allowed_model_types": [
"chat_text",
"chat_multimodal"
],
"configured": false,
"default_model_id": "",
"models": []
},
{
"slot_id": "agent_chat",
"name": "Agent chat",
"frontend_behavior": "selectable",
"allowed_model_types": [
"chat_text",
"chat_multimodal"
],
"configured": true,
"default_model_id": "model-example",
"models": [
{
"id": "model-example",
"name": "model-example",
"type": "chat_text",
"position": 0
}
],
"default_model_name": "model-example",
"extra_json": "{}"
}
]
}
Field |
Type |
Description |
|---|---|---|
|
object (array of objects) |
The current source registers 12 slots; see the shared page for their order and definitions. This API has no pagination. |
|
string |
Feature slot identifier. |
|
string |
Name in the currently registered definition. |
|
string |
hidden or selectable; describes whether the frontend allows users to choose a model. |
|
string (array of strings) |
Catalog model types allowed for assignment. |
|
boolean |
Whether a nonempty model list is configured; does not indicate model health or whether current credentials can call it. |
|
string |
Default model ID; an empty string when unconfigured. |
|
string |
Optional; currently displayed as the model ID. |
|
object (array of objects) |
Ordered model list; empty when unconfigured. |
|
string |
Candidate catalog model ID. |
|
string |
Currently projected name. |
|
string |
The list may supplement this from the catalog; write responses may currently use an empty string. |
|
integer |
Zero-based position in the list. |
|
string |
Optional extra JSON text, not a JSON object field. |
Error responses¶
{
"code": "ErrForbidden",
"msg": "Forbidden"
}
Field |
Type |
Description |
|---|---|---|
|
string |
Current error code. |
|
string |
Description of the current failure stage; this example uses HTTP Forbidden text. |
HTTP / code |
Conditions and handling |
|---|---|
400 / ErrParamInvalid |
Invalid slot, model, default, JSON text, or log-count parameter. |
401 / Unauthorized |
Invalid identity or an inapplicable authentication method such as X-API-Key. |
403 / ErrForbidden |
No Genesis administrator eligibility. |
503 / ErrServiceUnavailable |
Classified required identity/configuration dependencies are unavailable. |
500 / ErrServer or InternalServerError |
Internal or identity-stage failure. Read back after a write error rather than assuming rollback. |
Errors use code/msg, whereas success returns a raw object. Different failure stages may use different string codes; do not assume a distinct machine reason for every validation error.