curl --request POST \
--url https://paradigm.lighton.ai/api/v3/instance/ml-models \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"name": "<string>",
"technical_name": "<string>",
"litellm_model": "<string>",
"model_type": "<string>"
}
'import requests
url = "https://paradigm.lighton.ai/api/v3/instance/ml-models"
payload = {
"name": "<string>",
"technical_name": "<string>",
"litellm_model": "<string>",
"model_type": "<string>"
}
headers = {
"Authorization": "Bearer <token>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
body: JSON.stringify({
name: '<string>',
technical_name: '<string>',
litellm_model: '<string>',
model_type: '<string>'
})
};
fetch('https://paradigm.lighton.ai/api/v3/instance/ml-models', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));{
"name": "alfred-ft5",
"technical_name": "alfred-ft5",
"litellm_model": "hosted_vllm/alfred-ft5",
"model_type": "Large Language Model",
"endpoint": "http://models-gra11-afe6f177.lighton.ai:30424/v1/",
"max_model_len": 32768,
"enabled": true,
"default": false,
"parser": false,
"is_default_reranker": false,
"use_vllm_guided_choice": true
}Create a managed ML model (instance admin)
Register a new instance-wide managed model. Reserved to Sys Admin / Admin roles.
curl --request POST \
--url https://paradigm.lighton.ai/api/v3/instance/ml-models \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"name": "<string>",
"technical_name": "<string>",
"litellm_model": "<string>",
"model_type": "<string>"
}
'import requests
url = "https://paradigm.lighton.ai/api/v3/instance/ml-models"
payload = {
"name": "<string>",
"technical_name": "<string>",
"litellm_model": "<string>",
"model_type": "<string>"
}
headers = {
"Authorization": "Bearer <token>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
body: JSON.stringify({
name: '<string>',
technical_name: '<string>',
litellm_model: '<string>',
model_type: '<string>'
})
};
fetch('https://paradigm.lighton.ai/api/v3/instance/ml-models', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));{
"name": "alfred-ft5",
"technical_name": "alfred-ft5",
"litellm_model": "hosted_vllm/alfred-ft5",
"model_type": "Large Language Model",
"endpoint": "http://models-gra11-afe6f177.lighton.ai:30424/v1/",
"max_model_len": 32768,
"enabled": true,
"default": false,
"parser": false,
"is_default_reranker": false,
"use_vllm_guided_choice": true
}Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Body
Human-friendly display name.
Unique technical identifier used for routing.
Underlying LiteLLM provider routing string (e.g. hosted_vllm/my-model, openai/gpt-4o-mini). Distinct from technical_name — never assume they match.
Model type (e.g. Large Language Model, Embedding Model).
Optional explicit UUID; auto-generated when omitted.
Model serving endpoint URL. Omit for models with no custom endpoint.
Maximum context length in tokens. Omit if not applicable/known.
Whether the model is active.
Whether this is the default generation model.
Whether the model can be used as a parser.
Whether this is the default reranker.
Use vLLM guided_choice for constrained decoding.
Response
Unique ID for this model.
Display name.
Instance-wide unique technical name used for routing.
Model type.
Whether the model is active instance-wide.
Whether this is the platform default generation model.
Whether the model can be used as a parser.
Whether this is the default reranker.
Use vLLM guided_choice for constrained decoding.
Health of the model as of the last periodic gateway healthcheck. Read from cache — reading this endpoint never triggers a check.
Show child attributes
Show child attributes
Underlying LiteLLM provider routing string (e.g. hosted_vllm/my-model).
Model serving endpoint URL.
Maximum context length in tokens.
Deployment type (Self-Hosted or External).
Start messages template name.