> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pioneer.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Anthropic-compatible POST /v1/messages on Pioneer API

> Use Pioneer as a drop-in Anthropic SDK replacement. Point base_url to https://api.pioneer.ai/v1 and use your Pioneer API key to access fine-tuned models.

Pioneer implements the Anthropic Messages API so you can route existing Anthropic SDK code to your fine-tuned Pioneer models. Set `base_url` to `https://api.pioneer.ai/v1`, authenticate with your Pioneer API key, and use your training job ID as the model name. No other code changes are required.

## Endpoints

| Method | Path           | Description                             |
| ------ | -------------- | --------------------------------------- |
| `POST` | `/v1/messages` | Create a message (Anthropic-compatible) |

## Configure the Anthropic SDK

Pass your Pioneer API key and base URL when constructing the client:

<CodeGroup>
  ```python Python theme={null}
  import anthropic

  client = anthropic.Anthropic(
      api_key="YOUR_API_KEY",
      base_url="https://api.pioneer.ai/v1"
  )
  ```
</CodeGroup>

## Create a message

`POST /v1/messages` accepts the same request shape as the Anthropic Messages API. Set `model` to your training job ID and include your `messages` array. You can pass Pioneer-specific fields like `schema` alongside the standard Anthropic fields.

### Request parameters

<ParamField body="model" type="string" required>
  Training job ID (e.g. `job_abc123`) or a base model ID. This is the model that processes your request.
</ParamField>

<ParamField body="max_tokens" type="number" required>
  Maximum number of tokens to generate in the response.
</ParamField>

<ParamField body="messages" type="object[]" required>
  Conversation messages. Each object has a `role` (`"user"` or `"assistant"`) and a `content` string.
</ParamField>

<ParamField body="schema" type="object">
  Pioneer-specific extraction schema. Define `entities`, `classifications`, `structures`, or `relations` to control what the model extracts. See [Pioneer inference](/api-reference/inference/pioneer) for full schema documentation.
</ParamField>

### Examples

<CodeGroup>
  ```python Python (Anthropic SDK) theme={null}
  import anthropic

  client = anthropic.Anthropic(
      api_key="YOUR_API_KEY",
      base_url="https://api.pioneer.ai/v1"
  )

  message = client.messages.create(
      model="YOUR_TRAINING_JOB_ID",
      max_tokens=1024,
      messages=[
          {
              "role": "user",
              "content": "Extract entities from: Apple launched the iPhone."
          }
      ]
  )

  print(message.content[0].text)
  ```

  ```bash cURL theme={null}
  curl -X POST https://api.pioneer.ai/v1/messages \
    -H "X-API-Key: YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "YOUR_TRAINING_JOB_ID",
      "max_tokens": 1024,
      "messages": [
        {
          "role": "user",
          "content": "Extract entities from: Apple launched the iPhone."
        }
      ],
      "schema": {
        "entities": ["organization", "product"]
      }
    }'
  ```
</CodeGroup>

## Streaming

The `/v1/messages` endpoint supports streaming. Set `stream=True` in the SDK or `"stream": true` in the raw request body.

<CodeGroup>
  ```python Python (Anthropic SDK) theme={null}
  import anthropic

  client = anthropic.Anthropic(
      api_key="YOUR_API_KEY",
      base_url="https://api.pioneer.ai/v1"
  )

  with client.messages.stream(
      model="YOUR_TRAINING_JOB_ID",
      max_tokens=1024,
      messages=[{"role": "user", "content": "Summarize the following: ..."}]
  ) as stream:
      for text in stream.text_stream:
          print(text, end="", flush=True)
  ```

  ```bash cURL theme={null}
  curl -X POST https://api.pioneer.ai/v1/messages \
    -H "X-API-Key: YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "YOUR_TRAINING_JOB_ID",
      "max_tokens": 1024,
      "messages": [{"role": "user", "content": "Summarize the following: ..."}],
      "stream": true
    }'
  ```
</CodeGroup>

## Related

* [Pioneer native inference](/api-reference/inference/pioneer) — direct Pioneer endpoint with full schema documentation
* [OpenAI-compatible inference](/api-reference/inference/openai-compatible) — use the OpenAI SDK instead
* [Inference history and feedback](/api-reference/inference/history) — retrieve past results and submit corrections