curl --request POST \
--url https://pria.praxislxp.com/api/admin/aimodel \
--header 'Content-Type: application/json' \
--header 'x-access-token: <api-key>' \
--data '
{
"name": "<string>",
"institution": "<string>",
"model_use": "conversationModel",
"description": "<string>",
"client_library": "openai_cli",
"api_url": "<string>",
"api_key": "<string>",
"default_headers": {},
"features": {
"stream": false,
"tools": false,
"code": false,
"vision": false,
"mcp": false
},
"input_media": {
"text": true,
"image": false,
"audio": false,
"video": false,
"pdf": false
},
"output_media": {
"text": true,
"audio": false,
"video": false,
"image": false
},
"max_input_tokens": 128000,
"max_output_tokens": 64000,
"extended_context": false,
"status": "active",
"picture_url": "<string>"
}
'{
"success": true,
"id": "<string>",
"message": "AIModel created!"
}Create a new AI model
Creates a new AI model configuration. The handler validates the client_library / api_url / api_key combination via validateAIModelConfig before saving — openai_cli requires both api_url and api_key; mistral_cli and xai_cli require api_key; bedrock_cli requires neither (uses AWS credentials). When institution is omitted, the caller’s institution is used.
curl --request POST \
--url https://pria.praxislxp.com/api/admin/aimodel \
--header 'Content-Type: application/json' \
--header 'x-access-token: <api-key>' \
--data '
{
"name": "<string>",
"institution": "<string>",
"model_use": "conversationModel",
"description": "<string>",
"client_library": "openai_cli",
"api_url": "<string>",
"api_key": "<string>",
"default_headers": {},
"features": {
"stream": false,
"tools": false,
"code": false,
"vision": false,
"mcp": false
},
"input_media": {
"text": true,
"image": false,
"audio": false,
"video": false,
"pdf": false
},
"output_media": {
"text": true,
"audio": false,
"video": false,
"image": false
},
"max_input_tokens": 128000,
"max_output_tokens": 64000,
"extended_context": false,
"status": "active",
"picture_url": "<string>"
}
'{
"success": true,
"id": "<string>",
"message": "AIModel created!"
}Documentation Index
Fetch the complete documentation index at: https://docs.praxis-ai.com/llms.txt
Use this file to discover all available pages before exploring further.
Authorizations
JWT token passed in x-access-token header
Body
Model name (must be unique per institution + model_use).
Institution ID. Defaults to the caller's institution when omitted.
conversationModel, imageGenerationModel, audioGenerationModel, videoGenerationModel, imageAnalysisModel, embeddingsModel, audioAnalysisModel, ttsModel, summaryModel, rtModel, moderationModel SDK / dispatch backend used to talk to the provider.
openai_cli, anthropic_cli, bedrock_cli, google_cli, mistral_cli, xai_cli, stability_cli, lemonslice_cli Provider base URL. Required for openai_cli custom models.
Provider API key. Required for openai_cli, mistral_cli, and xai_cli custom models.
Show child attributes
Show child attributes
Capability flags advertised by the model.
Show child attributes
Show child attributes
Input modalities the model accepts.
Show child attributes
Show child attributes
Output modalities the model can produce.
Show child attributes
Show child attributes
none, low, medium, high, max active, inactive Was this page helpful?