### API Key Example

Source: https://platform.apodex.ai/docs

Your API key will look like this. Keep it secure!

```text
# Your API key will look like this:
sk_live_1234567890abcdef...
```

--------------------------------

### Fetch Response Snapshot (cURL)

Source: https://platform.apodex.ai/docs/responses-api

Example of retrieving a full response object snapshot using a GET request to the /v1/responses/:id endpoint.

```shell
curl https://api.apodex.ai/v1/responses/resp_<id> \
  -H "Authorization: Bearer YOUR_API_KEY"
```

--------------------------------

### SSE Stream Lifecycle Example

Source: https://platform.apodex.ai/docs/streaming

Illustrates the sequence of chunks emitted during a streaming response, from role assignment to the final [DONE] signal.

```text
1. role chunk        {"delta": {"role": "assistant"}}
2. reasoning chunks  {"delta": {"reasoning_steps": [...]}}  (thinking, tool use)
3. content chunks    {"delta": {"content": "..."}}          (final answer)
4. finish chunk      {"delta": {}, "finish_reason": "stop", "usage": {...}}
5. [DONE] signal     data: [DONE]
```

--------------------------------

### SSE Chunk Format Example

Source: https://platform.apodex.ai/docs/streaming

Demonstrates the structure of Server-Sent Events (SSE) data chunks, including content and final usage information.

```text
data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","model":"apodex-1-0-deepresearch-mini","created":1712345678,"choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}

data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","model":"apodex-1-0-deepresearch-mini","created":1712345678,"choices":[{"index":0,"delta":{},"finish_reason":"stop"}],"usage":{"prompt_tokens":10,"completion_tokens":100,"total_tokens":110}}

data: [DONE]
```

--------------------------------

### SSE Heartbeat Comment Example

Source: https://platform.apodex.ai/docs/streaming

Illustrates an SSE comment line used as a keep-alive heartbeat to prevent timeouts during idle periods.

```text
: heartbeat
```

--------------------------------

### SSE Event Stream Response Example

Source: https://platform.apodex.ai/docs/responses-api

Illustrates the structure of Server-Sent Events (SSE) frames received from the /v1/responses endpoint during a streaming request.

```text
event: response.created
data: {"type":"response.created","sequence_number":1,"response":{"id":"resp_<id>","status":"in_progress",...}}

event: response.in_progress
data: {"type":"response.in_progress","sequence_number":2,...}

event: response.output_item.added
data: {"type":"response.output_item.added","sequence_number":3,"output_index":0,"item":{"id":"rs_<id>","type":"reasoning","summary":[]}}

event: response.reasoning_text.delta
data: {"type":"response.reasoning_text.delta","sequence_number":4,"item_id":"rs_<id>","output_index":0,"content_index":0,"delta":"Thinking about ..."}

... (more reasoning + optional tool_call output items) ...

event: response.output_item.added
data: {"type":"response.output_item.added","sequence_number":N,"output_index":1,"item":{"id":"msg_<id>","type":"message","role":"assistant","content":[]}}

event: response.output_text.delta
data: {"type":"response.output_text.delta","sequence_number":N+1,"item_id":"msg_<id>","output_index":1,"content_index":0,"delta":"Recent advances ..."}

event: response.output_text.done
data: {"type":"response.output_text.done","sequence_number":N+M,...}

event: response.completed
data: {"type":"response.completed","sequence_number":N+M+1,"response":{"id":"resp_<id>","status":"completed","output":[...],"usage":{...}}}
```

--------------------------------

### Delta Object Example: Role Assignment

Source: https://platform.apodex.ai/docs/streaming

Shows the structure of a delta object when assigning a role to the assistant.

```json
{"delta": {"role": "assistant"}}
```

--------------------------------

### List Models API Response

Source: https://platform.apodex.ai/docs/chat-completions

Example JSON response when listing available models via the API. Shows model IDs, context lengths, and other metadata.

```json
{
  "object": "list",
  "data": [
    {
      "id": "apodex-1-0-deepresearch-h",
      "object": "model",
      "created": 1700000000,
      "owned_by": "apodex",
      "context_length": 131072,
      "max_completion_tokens": 262144
    },
    {
      "id": "apodex-1-0-deepresearch",
      "object": "model",
      "created": 1700000000,
      "owned_by": "apodex",
      "context_length": 262144,
      "max_completion_tokens": 16384
    },
    {
      "id": "apodex-1-0-deepresearch-mini",
      "object": "model",
      "created": 1700000000,
      "owned_by": "apodex",
      "context_length": 262144,
      "max_completion_tokens": 16384
    }
  ]
}
```

--------------------------------

### Resume SSE Stream (cURL)

Source: https://platform.apodex.ai/docs/responses-api

Example of resuming an SSE stream from a specific sequence number using the ?stream=true&after=SEQ query parameter.

```shell
curl https://api.apodex.ai/v1/responses/resp_<id>?stream=true&after=SEQ \
  -H "Authorization: Bearer YOUR_API_KEY"
```

--------------------------------

### GET /v1/responses

Source: https://platform.apodex.ai/docs/responses-api

Lists the authenticated user's recent responses. This endpoint returns lifecycle metadata only and does not include output. To fetch individual responses with their output, use GET /v1/responses/:id.

```APIDOC
## GET /v1/responses

### Description
List the authenticated user's recent responses. Lifecycle metadata only — no output[]; fetch individual responses via GET /v1/responses/:id.

### Method
GET

### Endpoint
/v1/responses

### Headers
- **Authorization** (string) - Required - Your Apodex API key (`Bearer YOUR_API_KEY`)

### Response
#### Success Response (200)
- A list of response metadata objects.
```

--------------------------------

### Handle SSE Stream Response

Source: https://platform.apodex.ai/docs

The response streams as SSE by default. Parse SSE lines to get reasoning steps and content.

```text
# Stream is the default. Parse SSE lines:
data: {"choices":[{"delta":{"content":"Hello!"}}]}
data: [DONE]
```

--------------------------------

### Submit Response via SSE Stream (cURL)

Source: https://platform.apodex.ai/docs/responses-api

Example of submitting a request to the /v1/responses endpoint using cURL to receive a Server-Sent Events (SSE) stream.

```shell
curl -N -X POST https://api.apodex.ai/v1/responses \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ 
    "model": "apodex-1-0-deepresearch-mini",
    "input": "Summarize the recent advances in fusion energy."
  }'
```

--------------------------------

### POST /v1/responses

Source: https://platform.apodex.ai/docs/responses-api

Creates a response. By default, it streams events via SSE. If stream=false, it blocks until the workflow completes and returns the full Response JSON. If background=true, the call returns immediately, and the result can be fetched later via GET /v1/responses/:id.

```APIDOC
## POST /v1/responses

### Description
Creates a response. With stream=true (default) the response body is an SSE event stream; with stream=false the connection blocks until the workflow finishes and returns the full Response JSON. With background=true the call returns immediately and you fetch the result via GET /v1/responses/:id.

### Method
POST

### Endpoint
/v1/responses

### Headers
- **Authorization** (string) - Required - Your Apodex API key (`Bearer YOUR_API_KEY`)
- **Content-Type** (string) - Required - Must be application/json (`application/json`)

### Parameters
#### Query Parameters
- **model** (string) - Required - Model ID, e.g. "apodex-1-0-deepresearch-mini".
- **input** (string | array) - Required - A user prompt as a plain string, or an array of {type, role, content} input items.
- **stream** (boolean) - Whether to stream events via SSE. Defaults to true.
- **background** (boolean) - If true, returns the Response object immediately (status="in_progress") without holding a connection. Use GET /v1/responses/:id (or ?stream=true to resume the SSE stream) to fetch the result later. Mutually exclusive with stream=true.
- **max_output_tokens** (integer) - Maximum number of tokens to generate.
- **temperature** (number) - Sampling temperature.
- **metadata** (object) - Free-form key-value metadata stored with the response.

### Request Example
```json
{
  "model": "apodex-1-0-deepresearch-mini",
  "input": "Summarize the recent advances in fusion energy."
}
```

### Response
#### Success Response (200)
- **event**: response.created
  - **data**: {"type":"response.created","sequence_number":1,"response":{"id":"resp_<id>","status":"in_progress",...}}
- **event**: response.output_text.delta
  - **data**: {"type":"response.output_text.delta","sequence_number":4,"item_id":"rs_<id>","output_index":0,"content_index":0,"delta":"Thinking about ..."}
- **event**: response.completed
  - **data**: {"type":"response.completed","sequence_number":N+M+1,"response":{"id":"resp_<id>","status":"completed","output":[...],"usage":{...}}}

#### Response Example
(Streaming SSE events as described above, or a JSON snapshot if stream=false)
```

--------------------------------

### Non-Streaming Chat Completion Response

Source: https://platform.apodex.ai/docs/chat-completions

Example of a non-streaming response from the Chat Completions API. Includes completion details and token usage.

```json
{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1712345678,
  "model": "apodex-1-0-deepresearch-mini",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Here are the latest trends in AI..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 256,
    "total_tokens": 268
  }
}
```

--------------------------------

### GET /v1/responses/:id

Source: https://platform.apodex.ai/docs/responses-api

Fetches a Response by its ID. By default, it returns a full Response object snapshot reconstructed from the workflow's event stream. With ?stream=true, it opens an SSE stream that replays the event history and continues live until the workflow terminates.

```APIDOC
## GET /v1/responses/:id

### Description
Fetch a Response by id. Default: returns the full Response object snapshot reconstructed from the workflow's event stream. With ?stream=true: opens an SSE stream that replays the entire event history through the translator and continues live until the workflow terminates — clients that lost their connection mid-stream can reconnect with ?stream=true&after=SEQ to skip events they already saw.

### Method
GET

### Endpoint
/v1/responses/:id

### Headers
- **Authorization** (string) - Required - Your Apodex API key (`Bearer YOUR_API_KEY`)

### Parameters
#### Query Parameters
- **stream** (boolean) - If true, returns SSE event stream (replay + live). If absent or false, returns the JSON snapshot.
- **after** (integer) - Only meaningful with stream=true. Skip events whose sequence_number is ≤ this value. Use to resume a stream where you left off.

### Response
#### Success Response (200)
- **id** (string) - The response ID.
- **object** (string) - The object type, always "response".
- **status** (string) - The status of the response (e.g., "completed").
- **model** (string) - The model ID used.
- **output** (array) - The reconstructed output of the response.
- **usage** (object) - Usage statistics.

#### Response Example
```json
{
  "id": "resp_<workflow_id>",
  "object": "response",
  "status": "completed",
  "model": "apodex-1-0-deepresearch-mini",
  "output": [...],
  "usage": {...}
}
```
```

--------------------------------

### List Models via API

Source: https://platform.apodex.ai/docs/chat-completions

Retrieve a list of available models and their capabilities programmatically. Useful for clients that need to enumerate or check model capabilities at runtime.

```bash
curl https://api.apodex.ai/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"
```

--------------------------------

### List Models

Source: https://platform.apodex.ai/docs/chat-completions

Retrieves a list of available models and their capabilities. Useful for clients to enumerate or check model capabilities at runtime.

```APIDOC
## GET /v1/models

### Description
Retrieves a list of available models and their capabilities.

### Method
GET

### Endpoint
/v1/models

### Response
#### Success Response (200)
- **object** (string) - Type of the response, "list".
- **data** (array) - An array of model objects.
  - **id** (string) - The model ID.
  - **object** (string) - Type of the object, "model".
  - **created** (integer) - Timestamp of model creation.
  - **owned_by** (string) - The organization that owns the model.
  - **context_length** (integer) - The maximum context length of the model.
  - **max_completion_tokens** (integer) - The maximum number of tokens the model can generate.

### Request Example
```bash
curl https://api.apodex.ai/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"
```
```

--------------------------------

### Chat Completions API

Source: https://platform.apodex.ai/docs

OpenAI-compatible chat completions with built-in reasoning, tool use, and streaming.

```APIDOC
## POST /v1/chat/completions

### Description
Send a POST request to the Chat Completions endpoint. Works with any OpenAI-compatible client.

### Method
POST

### Endpoint
https://api.apodex.ai/v1/chat/completions

### Parameters
#### Headers
- **Authorization** (string) - Required - Bearer YOUR_API_KEY
- **Content-Type** (string) - Required - application/json

### Request Body
- **model** (string) - Required - The model to use for chat completions.
- **messages** (array) - Required - An array of message objects, each with a role and content.
  - **role** (string) - Required - The role of the author of the message (e.g., "user", "assistant").
  - **content** (string) - Required - The content of the message.

### Request Example
```json
{
  "model": "apodex-1-0-deepresearch-mini",
  "messages": [
    {
      "role": "user",
      "content": "Hello!"
    }
  ]
}
```

### Response
#### Success Response (200)
- **choices** (array) - An array of choices, each containing a delta with content.
  - **delta** (object) - The change in the assistant's response.
    - **content** (string) - The content of the message chunk.
```

--------------------------------

### OpenAI SDK Streaming Chat Completions (JavaScript)

Source: https://platform.apodex.ai/docs/streaming

Consumes SSE streams from the Chat Completions API using the official OpenAI SDK in JavaScript. Handles content streaming and final finish reason/usage.

```javascript
import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.APODEX_API_KEY,
  baseURL: 'https://api.apodex.ai/v1',
});

const stream = await client.chat.completions.create({
  model: 'apodex-1-0-deepresearch-mini',
  messages: [{ role: 'user', content: 'Hello' }],
  stream: true,
});

for await (const chunk of stream) {
  const choice = chunk.choices[0];
  const delta = choice.delta;

  // Final answer (token-by-token)
  if (delta?.content) {
    process.stdout.write(delta.content);
  }

  // Finish + usage (last chunk)
  if (choice.finish_reason) {
    console.log(`\nFinish: ${choice.finish_reason}`);
    if (chunk.usage) console.log('Usage:', chunk.usage);
  }
}
```

--------------------------------

### Create Chat Completion (cURL)

Source: https://platform.apodex.ai/docs/chat-completions

Send a conversation to the model and receive a completion using cURL. Supports streaming responses.

```bash
curl -X POST https://api.apodex.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "apodex-1-0-deepresearch-mini",
    "messages": [
      {
        "role": "user",
        "content": "What are the latest trends in AI?"
      }
    ],
    "stream": true
  }'
```

--------------------------------

### Authenticate API Request with JavaScript

Source: https://platform.apodex.ai/docs/authentication

This JavaScript snippet shows how to authenticate API requests using environment variables for your API key. Ensure your API key is stored securely.

```javascript
const response = await fetch('https://api.apodex.ai/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Authorization': `Bearer ${process.env.APODEX_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    messages: [{
      role: 'user',
      content: 'Hello!'
    }]
  })
});
```

--------------------------------

### Authenticate API Request with cURL

Source: https://platform.apodex.ai/docs/authentication

Use this cURL command to authenticate your API requests by including your API key in the Authorization header.

```bash
curl https://api.apodex.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json"
```

--------------------------------

### Create Chat Completion

Source: https://platform.apodex.ai/docs/chat-completions

Sends a conversation to the model and receives a completion. Supports streaming (SSE) and non-streaming modes.

```APIDOC
## POST /v1/chat/completions

### Description
Sends a conversation to the model and receives a completion. Supports streaming (SSE) and non-streaming modes.

### Method
POST

### Endpoint
/v1/chat/completions

### Parameters
#### Headers
- **Authorization** (string) - Required. Your Apodex API key. Example: `Bearer YOUR_API_KEY`
- **Content-Type** (string) - Required. Must be `application/json`.

#### Request Body
- **model** (string) - Required. Model ID to use, e.g. "apodex-1-0-deepresearch-mini".
- **messages** (array) - Required. Array of message objects with "role" (system/user/assistant) and "content" fields.
- **stream** (boolean) - Optional. Whether to stream the response via SSE. Defaults to true.
- **max_tokens** (integer) - Optional. Maximum number of tokens to generate in the completion.
- **mcp_servers** (array) - Optional. Array of MCP server configs ({name, url, headers?, access_token?, oauth?}) for external tool access.

### Request Example
```bash
curl -X POST https://api.apodex.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d 
    {
      "model": "apodex-1-0-deepresearch-mini",
      "messages": [
        {
          "role": "user",
          "content": "What are the latest trends in AI?"
        }
      ],
      "stream": true
    }
```

### Response
#### Success Response (Non-Streaming)
- **id** (string) - Unique identifier for the completion.
- **object** (string) - Type of the object, "chat.completion".
- **created** (integer) - Timestamp of completion creation.
- **model** (string) - The model used for the completion.
- **choices** (array) - An array of completion choices.
  - **index** (integer) - Index of the choice.
  - **message** (object) - The message content.
    - **role** (string) - Role of the message sender (assistant).
    - **content** (string) - The generated text content.
  - **finish_reason** (string) - The reason the model stopped generating tokens.
- **usage** (object) - Token usage statistics.
  - **prompt_tokens** (integer) - Number of tokens in the prompt.
  - **completion_tokens** (integer) - Number of tokens in the completion.
  - **total_tokens** (integer) - Total tokens used.

#### Response Example (Non-Streaming)
```json
{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1712345678,
  "model": "apodex-1-0-deepresearch-mini",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Here are the latest trends in AI..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 256,
    "total_tokens": 268
  }
}
```

#### Response Example (Streaming)
When `stream: true`, the response is delivered as Server-Sent Events (SSE). Each line is prefixed with `data: ` followed by a JSON chunk. The stream ends with `data: [DONE]`.

**Phase 1: Reasoning**
- `delta.reasoning_steps` (array) - Emits reasoning steps with `type` (e.g., thinking, web_search) and `content`.

**Phase 2: Final Answer**
- `delta.content` (string) - Streams the final answer token by token.
```

--------------------------------

### Send a Chat Completions Request

Source: https://platform.apodex.ai/docs

Send a POST request to the Chat Completions endpoint. Works with any OpenAI-compatible client.

```curl
curl -X POST https://api.apodex.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d {"model": "apodex-1-0-deepresearch-mini", "messages": [{"role": "user", "content": "Hello!"}]}
```

--------------------------------

### Response Snapshot JSON

Source: https://platform.apodex.ai/docs/responses-api

The JSON structure returned when fetching a response snapshot (stream=false) from the /v1/responses/:id endpoint.

```json
{
  "id": "resp_<workflow_id>",
  "object": "response",
  "status": "completed",
  "model": "apodex-1-0-deepresearch-mini",
  "output": [...],
  "usage": {...}
}
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.