# Chat Completions API

> POST /v1/chat/completions — the OpenAI-format endpoint, for Cursor, LangChain, LiteLLM and the OpenAI SDKs.

_Source: https://aiprimetech.io/docs/api-reference/chat-completions/ · Home > Docs > API reference_

The OpenAI-compatible surface. Same models, same prices, different envelope.

- **Endpoint:** `POST https://aiprimetech.io/v1/chat/completions`
- **Base URL for SDKs:** `https://aiprimetech.io/v1` — note the /v1
- **Auth:** `Authorization: Bearer sk-...`


## Request body

| Field | Type | Required | Description |
|---|---|---|---|
| `model` | string | yes | Any id from `/v1/models`, including Claude ids |
| `messages` | array | yes | Supports system, user and assistant roles |
| `max_tokens` | integer | no | Optional here, unlike the Messages API |
| `temperature` | number | no | 0–2 in the OpenAI convention |
| `stream` | boolean | no | true for SSE chunks |
| `tools` | array | no | OpenAI function-calling shape |

> Unlike Messages, `system` **is** a role here — put it as the first entry of `messages`. Do not send a top-level `system` field to this endpoint.


## Example

```bash
curl https://aiprimetech.io/v1/chat/completions \
  -H "Authorization: Bearer $CLAUDEAPIKEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "messages": [{"role": "user", "content": "Hello"}]
  }'
```


## Response

```json
{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "claude-sonnet-4-6",
  "choices": [{
    "index": 0,
    "message": {"role": "assistant", "content": "..."},
    "finish_reason": "stop"
  }],
  "usage": {"prompt_tokens": 9, "completion_tokens": 12, "total_tokens": 21}
}
```


## Using the OpenAI SDK

```python
from openai import OpenAI

client = OpenAI(
    api_key="sk-your-key",
    base_url="https://aiprimetech.io/v1",   # the /v1 matters
)

resp = client.chat.completions.create(
    model="claude-sonnet-4-6",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)
```


## Field mapping between the two formats

| Concept | Messages | Chat Completions |
|---|---|---|
| System prompt | top-level `system` | `messages[0]` with role system |
| Reply text | `content[0].text` | `choices[0].message.content` |
| Why it stopped | `stop_reason` | `finish_reason` |
| Input tokens | `usage.input_tokens` | `usage.prompt_tokens` |
| Output tokens | `usage.output_tokens` | `usage.completion_tokens` |
| Stream delta | `content_block_delta` | `choices[0].delta.content` |

- [Cursor](https://aiprimetech.io/docs/guides/cursor/) — Configure Cursor against this endpoint
- [Migrating from OpenAI](https://aiprimetech.io/docs/guides/migration-from-openai/) — Switch an existing codebase
- [Streaming](https://aiprimetech.io/docs/api-reference/streaming/) — SSE for both formats


---
_ClaudeAPIKey.dev is an independently operated, Anthropic-compatible API gateway. Not affiliated with Anthropic._