Chat Completions Streaming API

OpenAI Compatible Endpoint

Overview

The Chat Completions API provides an OpenAI-compatible interface for interacting with the Pulse AI Research Agent. This endpoint enables real-time streaming responses using Server-Sent Events (SSE), allowing applications to display AI-generated content as it's being created.

The API follows the OpenAI chat completions format for maximum compatibility with existing tools and libraries designed for OpenAI's API.

Base URL

https://insight.lineupiq.io/v1/chat/completions

Authentication

The API requires authentication using both an API key and an application ID:

Header	Description	Required
`x-api-key`	Your API key for authentication	Yes
`x-app-id`	Your application ID	Yes
`Content-Type`	Must be `application/json`	Yes

Authentication errors will return a 401 status code with details about the missing or invalid credentials.

Request Format

HTTP Method

POST

Request Body

{
  "model": "control-4v1",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant specializing in music industry knowledge."
    },
    {
      "role": "user",
      "content": "I'm looking for venues in Edmonton that can host a rock concert."
    }
  ],
  "stream": true
}

Parameters

Parameter	Type	Description	Required	Default
`model`	string	ID of the model to use (e.g., 'control-4v1', 'discovery-pulse-4v1')	Yes	-
`messages`	array	A list of messages comprising the conversation so far	Yes	-
`messages[].role`	string	The role of the message author (system, user, or assistant)	Yes	-
`messages[].content`	string	The content of the message	Yes	-
`stream`	boolean	Whether to stream the response (recommended)	No	true

Response Format

The API returns a stream of events in the Server-Sent Events (SSE) format with content type text/event-stream. Each event starts with data: followed by a JSON string and two newlines.

Successful Response (200 OK)

The streaming response consists of multiple events:

First Content Chunk

data: {"choices": [{"delta": {"role": "assistant", "content": "Hello"}, "index": 0, "finish_reason": null}]}\n\n

Subsequent Content Chunks

data: {"choices": [{"delta": {"content": ", I am your assistant"}, "index": 0, "finish_reason": null}]}\n\n

End Marker with Finish Reason

data: {"choices":[{"delta":{},"finish_reason":"stop","index":0}]}\n\n

Final Done Marker

data: [DONE]\n\n

Error Responses

Authentication Errors (401 Unauthorized)

Missing API Key:

{
  "detail": "API key missing. Provide it via X-API-Key header or Bearer token"
}

Invalid API Key:

{
  "detail": "Invalid API key"
}

Missing User Email:

{
  "error": {
    "message": "Authentication required",
    "type": "authentication_error",
    "param": null,
    "code": "invalid_request_error"
  }
}

In-Stream Errors

Errors that occur after the stream has started will be sent as events in the stream:

data: {"error": {"message": "An internal server error occurred", "type": "server_error", "param": null, "code": "internal_error"}}\n\n

Client Implementation

JavaScript Example

Python Example

Rate Limits

Plan	Requests per minute	Tokens per minute
Free	TBD	TBD
Basic	TBD	TBD
Pro	TBD	TBD
Enterprise	Custom	Custom

Exceeding these limits will result in a 429 Too Many Requests response.

Models

The following models are available through this endpoint:

Model ID	Description	Context Window	Use Case
discovery-pulse-4v1	Research-focused AI assistant	32K tokens	Deep research, complex queries

Best Practices

Always use streaming: Set stream: true for the best user experience, allowing content to appear incrementally.

Include conversation history: Pass the full conversation history in the messages array to maintain context.

Use system messages: Set the assistant's behavior and capabilities with a system message at the beginning of your messages array.

Handle errors gracefully: Implement proper error handling for both pre-stream and in-stream errors.

Implement reconnection logic: Have your client automatically reconnect if the connection is lost during streaming.

Limitations

Maximum request size: 32KB

Maximum conversation length: 32K tokens (varies by model)

Maximum streaming duration: 5 minutes

Support

If you encounter any issues or have questions about the Chat Completions API, please contact our support team at api-support@lineupiq.io.

Chat Completions API

Chat Completions Streaming API#

OpenAI Compatible Endpoint#

Overview#

Base URL#

Authentication#

Request Format#

HTTP Method#

Request Body#

Parameters#

Response Format#

Successful Response (200 OK)#

Error Responses#

Authentication Errors (401 Unauthorized)#

In-Stream Errors#

Client Implementation#

JavaScript Example#

Python Example#

Rate Limits#

Models#

Best Practices#

Limitations#

Support#