Continue text from a prompt. For chat conversations, use Chat Completions rather than sending messages directly to this endpoint.
Request
| Item | Description |
|---|---|
| Method | POST |
| Path | /v1/completions |
| Gateway URL | https://ailink.nativex.com |
| Authentication | Authorization: Bearer sk-xxxx; replace the placeholder with your API Key |
| Request format | application/json |
Replace MODEL_ID with a model from the Model Marketplace that supports this endpoint. Available models, parameters, and permissions depend on your account.
Parameters
| Parameter | Location | Type | Required | Description |
|---|---|---|---|---|
model |
Body | string | Yes | A model that supports legacy Completions |
prompt |
Body | string / string[] | Yes | Input prompt |
max_tokens |
Body | integer | No | Positive integer limiting output tokens |
temperature |
Body | number | No | Sampling temperature; allowed values depend on the model |
top_p |
Body | number | No | Nucleus sampling parameter |
stream |
Body | boolean | No | Whether to enable SSE |
stop |
Body | string / string[] | No | Stop sequences |
n |
Body | integer | No | Number of candidates. Start with 1; the supported range depends on the model. |
echo |
Body | boolean | No | Whether to echo the input; requires model support |
suffix |
Body | string | No | Suffix after the completion; requires model support |
Example
curl -X POST 'https://ailink.nativex.com/v1/completions' \
-H 'Authorization: Bearer sk-xxxx' \
-H 'Content-Type: application/json' \
-d '{
"model": "MODEL_ID",
"prompt": "Continue this sentence: The first rays of morning sunlight",
"max_tokens": 128
}'
Response
The following is a simplified response for readability. IDs, URLs, text, and usage values are illustrative.
{
"id": "cmpl_example",
"object": "text_completion",
"choices": [
{
"text": "lit up the desk by the window.",
"index": 0,
"finish_reason": "stop"
}
]
}
| Field or result | Description |
|---|---|
| choices[].text | Completed text |
| choices[].finish_reason | Reason generation stopped |
Error handling
For unsuccessful responses, check the HTTP status and error body first. Verify the API Key, model name, account balance, and parameters. If rate limited, reduce concurrency and retry later.