Create a model in the catalogue

curl --request POST \
  --url https://{host}/v2/workspaces/slug:llm-gateway/webhooks/v1/models \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model_id": "<string>"
}
'

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({model_id: '<string>'})
};

fetch('https://{host}/v2/workspaces/slug:llm-gateway/webhooks/v1/models', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

import requests

url = "https://{host}/v2/workspaces/slug:llm-gateway/webhooks/v1/models"

payload = { "model_id": "<string>" }
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

{
  "model_id": "<string>",
  "type": "completion",
  "display": {
    "name": "<string>",
    "brand": "<string>",
    "hidden": true
  },
  "capabilities": {
    "vision": true,
    "audio": true,
    "text": true,
    "image": true
  },
  "limits": {},
  "failover": "<string>",
  "region": "world",
  "end_of_life": "2023-12-25",
  "dimensions": "<unknown>",
  "supported_dimensions": [
    123
  ],
  "metrics": "<unknown>",
  "provider_config": {
    "exclude_params": [
      "<string>"
    ],
    "rename_params": {},
    "api": "chat_completions"
  },
  "pricing": {
    "input_per_1m_tokens": 123,
    "output_per_1m_tokens": 123,
    "cache_read_per_1m_tokens": 123,
    "cache_write_per_1m_tokens": 123
  },
  "tags": [
    "<string>"
  ],
  "org_slugs": [
    "<string>"
  ],
  "enabled": true
}

{
  "error": "<string>",
  "message": "<string>",
  "code": "<string>",
  "details": {},
  "status": 123,
  "retryAfter": 123,
  "provider": "<string>",
  "model": "<string>",
  "provider_error_type": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "code": "<string>",
  "details": {},
  "status": 123,
  "retryAfter": 123,
  "provider": "<string>",
  "model": "<string>",
  "provider_error_type": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "code": "<string>",
  "details": {},
  "status": 123,
  "retryAfter": 123,
  "provider": "<string>",
  "model": "<string>",
  "provider_error_type": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "code": "<string>",
  "details": {},
  "status": 123,
  "retryAfter": 123,
  "provider": "<string>",
  "model": "<string>",
  "provider_error_type": "<string>"
}

POST

models

Create a model in the catalogue

curl --request POST \
  --url https://{host}/v2/workspaces/slug:llm-gateway/webhooks/v1/models \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model_id": "<string>"
}
'

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({model_id: '<string>'})
};

fetch('https://{host}/v2/workspaces/slug:llm-gateway/webhooks/v1/models', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

import requests

url = "https://{host}/v2/workspaces/slug:llm-gateway/webhooks/v1/models"

payload = { "model_id": "<string>" }
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

{
  "model_id": "<string>",
  "type": "completion",
  "display": {
    "name": "<string>",
    "brand": "<string>",
    "hidden": true
  },
  "capabilities": {
    "vision": true,
    "audio": true,
    "text": true,
    "image": true
  },
  "limits": {},
  "failover": "<string>",
  "region": "world",
  "end_of_life": "2023-12-25",
  "dimensions": "<unknown>",
  "supported_dimensions": [
    123
  ],
  "metrics": "<unknown>",
  "provider_config": {
    "exclude_params": [
      "<string>"
    ],
    "rename_params": {},
    "api": "chat_completions"
  },
  "pricing": {
    "input_per_1m_tokens": 123,
    "output_per_1m_tokens": 123,
    "cache_read_per_1m_tokens": 123,
    "cache_write_per_1m_tokens": 123
  },
  "tags": [
    "<string>"
  ],
  "org_slugs": [
    "<string>"
  ],
  "enabled": true
}

{
  "error": "<string>",
  "message": "<string>",
  "code": "<string>",
  "details": {},
  "status": 123,
  "retryAfter": 123,
  "provider": "<string>",
  "model": "<string>",
  "provider_error_type": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "code": "<string>",
  "details": {},
  "status": 123,
  "retryAfter": 123,
  "provider": "<string>",
  "model": "<string>",
  "provider_error_type": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "code": "<string>",
  "details": {},
  "status": 123,
  "retryAfter": 123,
  "provider": "<string>",
  "model": "<string>",
  "provider_error_type": "<string>"
}

{
  "error": "<string>",
  "message": "<string>",
  "code": "<string>",
  "details": {},
  "status": 123,
  "retryAfter": 123,
  "provider": "<string>",
  "model": "<string>",
  "provider_error_type": "<string>"
}

Authorizations

Authorization

string

header

required

User-bound credential carrying an identity: either a session JWT or a user access token (at:*) generated from the user settings UI. Send as Authorization: Bearer <token>. Org API keys (iak_*) are not accepted here - they carry no user identity. Use the x-prismeai-api-key header instead (see OrgApiKeyAuth).

Body

application/json

Write-shape for POST /v1/models and each item of PUT /v1/models. model_id and type are required.

model_id

string

required

Maximum string length: 128

type

enum<string>

required

Available options:

completion,

embeddings,

image_generation,

transcription

Maximum string length: 64

display

object

capabilities

object

limits

object

failover

string

Maximum string length: 128

region

enum<string> | null

Available options:

world,

eu-west,

eu-north,

us-east,

us-west,

asia,

null

Maximum string length: 64

end_of_life

string<date> | null

dimensions

any

supported_dimensions

any

metrics

any

provider_config

any

pricing

any

Response

Model created. Returns the inserted document.

Model catalogue document. The catalogue is the source of truth for provider routing, capabilities, pricing, and per-org availability.

model_id

string

required

Stable identifier (matches the value sent in request.model).

Maximum string length: 128

type

enum<string>

required

Model family.

Available options:

completion,

embeddings,

image_generation,

transcription

Maximum string length: 64

display

object

Display metadata (label, brand, hidden flag).

Hide child attributes

display.name

string

display.brand

string

display.hidden

boolean

capabilities

object

Flags advertising what the model supports.

Hide child attributes

capabilities.vision

boolean

capabilities.audio

boolean

capabilities.text

boolean

capabilities.image

boolean

limits

object

Provider-side limits (e.g. context window, max tokens).

failover

string

Optional model_id to route to when the primary fails.

Maximum string length: 128

region

enum<string> | null

Optional carbon-region override. null (default) = the region the provider entry serving the model gives: its carbon_region, else the Bedrock inference-profile prefix (global. → world, eu. → eu-west…) or AWS region, else the Vertex location. Set it only for a model that runs elsewhere. An unknown value is refused (400 INVALID_REGION).

Available options:

world,

eu-west,

eu-north,

us-east,

us-west,

asia,

null

Maximum string length: 64

end_of_life

string<date> | null

Announced end-of-life date of the model (YYYY-MM-DD). Informational, routing continues.

dimensions

any

Default embedding dimensionality (embeddings models only).

supported_dimensions

number[]

Allowed values for the request dimensions parameter.

metrics

any

Free-form metrics block (latency, throughput hints, …).

provider_config

object

Provider-specific configuration (batch size, parallelism, …).

Hide child attributes

provider_config.exclude_params

string[]

Request-body parameters dropped for this model (on top of the provider's own exclusions), e.g. temperature on a reasoning model.

provider_config.rename_params

object

Request-body parameters renamed for this model, e.g. {"max_tokens": "max_completion_tokens"}.

Hide child attributes

provider_config.rename_params.{key}

string

provider_config.api

enum<string>

Upstream API the model is called with. Absent = chat_completions. responses is needed by models that refuse function tools with a reasoning_effort on Chat Completions (GPT-6 family). Prefilled from the catalogue when the model is created, never changed by the catalogue sync.

Available options:

chat_completions,

responses

pricing

object

Cost configuration used to compute usage.cost.

Hide child attributes

pricing.input_per_1m_tokens

number<double>

pricing.output_per_1m_tokens

number<double>

pricing.cache_read_per_1m_tokens

number<double>

Price of 1M input tokens read from the provider's prompt cache (0 = free). Optional: defaults to the input price × the provider ratio (claude 0.1, gpt-4o 0.5, gpt-4.1/o3/o4 0.25, others 0.1).

pricing.cache_write_per_1m_tokens

number<double>

Price of 1M input tokens written to the provider's prompt cache (Anthropic, 0 = free). Optional: defaults to the input price × 1.25 for claude, × 1 otherwise.

Introduction

Platform API

Agent Creator API

LLM Gateway API

Governance API

Knowledges API

Collections API

Create a model in the catalogue

Authorizations

Body

Response