Get a model
β οΈ
Preview β this feature is being built; the design below is the spec.
GET https://app.syntic.ai/v1/models/{model_id}Return one model by its exact name. Model names contain spaces, so URL-encode them (Amara%20S1.2).
Headers
| Header | Required | Value |
|---|---|---|
Authorization | Yes | Bearer <key>: a private inference key (sk_live_...). See Authentication. |
Path parameters
| Parameter | Type | Description |
|---|---|---|
model_id | string | The modelβs id, URL-encoded, for example Amara%20Quick. |
Response
200 OK, Content-Type: application/json
{
"id": "Amara Quick",
"object": "model",
"created": 1767225600,
"display_name": "Amara Quick",
"family": "amara",
"description": "Short, frequent exchanges where a fast answer and a low bill matter most.",
"default": false,
"available": true,
"context_window": 262144,
"max_output_tokens": 32768,
"capabilities": {
"tools": true,
"parallel_tool_calls": true,
"strict_tools": true,
"json_object": true,
"json_schema": true,
"vision": true,
"documents": true,
"citations": false,
"reasoning": false,
"temperature": true,
"top_p": true,
"top_k": false,
"penalties": false,
"seed": false,
"server_tools": [
"syntic_web_search",
"syntic_web_fetch"
]
},
"pricing": {
"currency": "USD",
"unit": "1M tokens",
"input": 1.5,
"cached_input": 0.25,
"output": 8.0,
"batch_discount": 0.5
},
"zero_data_retention": true,
"deprecation": null
}| Field | Type | Description |
|---|---|---|
id | string | The exact Syntic model name accepted in every model field. |
object | string | Always model. |
created | integer | Unix seconds when the model was added to the catalog. |
display_name | string | The name shown in Syntic products. Same as id today. |
family | string | amara or syntic-code. |
default | boolean | true for the model used when a chat completion omits model. |
available | boolean | Whether your plan includes the model. |
context_window | integer | Prompt plus reply, in tokens. |
max_output_tokens | integer | Highest accepted max_tokens. |
capabilities | object | Feature flags. Using a feature whose flag is false returns 400 naming the feature. |
pricing | object | Usage-credit rates per million tokens. Plan-covered requests carry no per-token charge. |
zero_data_retention | boolean | Usable by organizations in zero-data-retention mode. |
deprecation | object or null | announced_at, retires_at and replacement once a retirement is scheduled. |
Examples
curl -sS https://app.syntic.ai/v1/models/Amara%20Quick \
-H "Authorization: Bearer $SYNTIC_API_KEY"Errors
Errors use the error envelope and carry a request_id. Codes specific to this operation:
| Status | type | code | When |
|---|---|---|---|
| 404 | not_found_error | model_not_found | No model has that name. Names are matched exactly, including spaces and capitals. |
| 401 | authentication_error | invalid_api_key | The key is missing, unknown, revoked or expired. |
| 429 | rate_limit_error | rate_limited | Too many requests; wait for Retry-After. See Rate-limit headers. |
Unlike chat completions
Chat completions currently answers an unknown model name with the default model (not live yet β on develop: it returns 404 model_not_found too). This endpoint does not guess: an unknown name returns 404, which makes it a reliable way to validate model names in configuration at start-up.
Related
Was this page helpful?