Chat Completions

POST/v1/chat/completions

Creates a chat completion. This endpoint is fully compatible with the OpenAI Chat Completions API format.

You can use this API in two modes:

With Assistant: Provide assistant_id to use a pre-configured assistant with system prompt, model, and tools from the dashboard
Direct Mode: Provide model directly for a pure OpenAI-compatible experience without assistant features

Request Body

Prop

Type

Messages Format

Each message in the messages array has the following structure:

Prop

Type

Example Request (With Assistant)

Use an assistant to leverage pre-configured system prompts, model settings, and tools:

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.assistantrouter.com/v1',
  apiKey: process.env.ASSISTANTROUTER_API_KEY,
});

const response = await client.chat.completions.create({
  // @ts-ignore - assistant_id is our extension
  assistant_id: 'your-assistant-uuid',
  messages: [
    { role: 'user', content: 'What is the capital of France?' },
  ],
});

console.log(response.choices[0].message.content);

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.assistantrouter.com/v1",
    api_key=os.environ["ASSISTANTROUTER_API_KEY"],
)

response = client.chat.completions.create(
    extra_body={"assistant_id": "your-assistant-uuid"},
    messages=[
        {"role": "user", "content": "What is the capital of France?"},
    ],
)

print(response.choices[0].message.content)

curl https://api.assistantrouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $ASSISTANTROUTER_API_KEY" \
  -d '{
    "assistant_id": "your-assistant-uuid",
    "messages": [
      {"role": "user", "content": "What is the capital of France?"}
    ]
  }'

Example Request (Direct Mode)

Use the API directly without an assistant - pure OpenAI compatibility:

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.assistantrouter.com/v1',
  apiKey: process.env.ASSISTANTROUTER_API_KEY,
});

const response = await client.chat.completions.create({
  model: 'anthropic/claude-haiku-4.5',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'What is the capital of France?' },
  ],
});

console.log(response.choices[0].message.content);

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.assistantrouter.com/v1",
    api_key=os.environ["ASSISTANTROUTER_API_KEY"],
)

response = client.chat.completions.create(
    model="anthropic/claude-haiku-4.5",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "What is the capital of France?"},
    ],
)

print(response.choices[0].message.content)

curl https://api.assistantrouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $ASSISTANTROUTER_API_KEY" \
  -d '{
    "model": "anthropic/claude-haiku-4.5",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "What is the capital of France?"}
    ]
  }'

In direct mode, RAG (file_search) and web search tools are not available. Use an assistant to access these features.

Response

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1704067200,
  "model": "anthropic/claude-haiku-4.5",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "The capital of France is Paris."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 25,
    "completion_tokens": 8,
    "total_tokens": 33
  },
  "x_assistantrouter": {
    "cost_cents": 0.15,
    "assistant_id": "your-assistant-uuid",
    "conversation_id": "conv_abc123",
    "tools_used": [],
    "was_fallback": false
  }
}

The x_assistantrouter extension includes:

cost_cents: The cost of this request in cents
assistant_id: The assistant ID used for this request
conversation_id: The conversation ID (if provided or auto-created)
tools_used: Array of tools that were invoked (e.g., ["web_search"])
was_fallback: Whether a fallback model was used

Streaming

Set stream: true to receive partial responses as Server-Sent Events (SSE):

const stream = await client.chat.completions.create({
  // @ts-ignore
  assistant_id: 'your-assistant-uuid',
  messages: [{ role: 'user', content: 'Tell me a story.' }],
  stream: true,
});

for await (const chunk of stream) {
  const content = chunk.choices[0]?.delta?.content || '';
  process.stdout.write(content);
}

stream = client.chat.completions.create(
    extra_body={"assistant_id": "your-assistant-uuid"},
    messages=[{"role": "user", "content": "Tell me a story."}],
    stream=True,
)

for chunk in stream:
    content = chunk.choices[0].delta.content or ""
    print(content, end="", flush=True)

Tool Calling

Enable function calling by providing tools in your request:

const response = await client.chat.completions.create({
  // @ts-ignore
  assistant_id: 'your-assistant-uuid',
  messages: [{ role: 'user', content: 'What is the weather in Paris?' }],
  tools: [
    {
      type: 'function',
      function: {
        name: 'get_weather',
        description: 'Get the current weather for a location',
        parameters: {
          type: 'object',
          properties: {
            location: {
              type: 'string',
              description: 'The city and country',
            },
          },
          required: ['location'],
        },
      },
    },
  ],
});

Errors

Code	Description
`authentication_error`	Invalid or missing API key
`insufficient_balance`	Not enough credits in wallet
`rate_limit_exceeded`	Request rate limit exceeded
`not_found`	Assistant not found
`bad_request`	Invalid request parameters

Chat Completions

POST/v1/chat/completions

Creates a chat completion. This endpoint is fully compatible with the OpenAI Chat Completions API format.

You can use this API in two modes:

With Assistant: Provide assistant_id to use a pre-configured assistant with system prompt, model, and tools from the dashboard
Direct Mode: Provide model directly for a pure OpenAI-compatible experience without assistant features

Request Body

Prop

Type

Messages Format

Each message in the messages array has the following structure:

Prop

Type

Example Request (With Assistant)

Use an assistant to leverage pre-configured system prompts, model settings, and tools:

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.assistantrouter.com/v1',
  apiKey: process.env.ASSISTANTROUTER_API_KEY,
});

const response = await client.chat.completions.create({
  // @ts-ignore - assistant_id is our extension
  assistant_id: 'your-assistant-uuid',
  messages: [
    { role: 'user', content: 'What is the capital of France?' },
  ],
});

console.log(response.choices[0].message.content);

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.assistantrouter.com/v1",
    api_key=os.environ["ASSISTANTROUTER_API_KEY"],
)

response = client.chat.completions.create(
    extra_body={"assistant_id": "your-assistant-uuid"},
    messages=[
        {"role": "user", "content": "What is the capital of France?"},
    ],
)

print(response.choices[0].message.content)

curl https://api.assistantrouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $ASSISTANTROUTER_API_KEY" \
  -d '{
    "assistant_id": "your-assistant-uuid",
    "messages": [
      {"role": "user", "content": "What is the capital of France?"}
    ]
  }'

Example Request (Direct Mode)

Use the API directly without an assistant - pure OpenAI compatibility:

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.assistantrouter.com/v1',
  apiKey: process.env.ASSISTANTROUTER_API_KEY,
});

const response = await client.chat.completions.create({
  model: 'anthropic/claude-haiku-4.5',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'What is the capital of France?' },
  ],
});

console.log(response.choices[0].message.content);

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.assistantrouter.com/v1",
    api_key=os.environ["ASSISTANTROUTER_API_KEY"],
)

response = client.chat.completions.create(
    model="anthropic/claude-haiku-4.5",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "What is the capital of France?"},
    ],
)

print(response.choices[0].message.content)

curl https://api.assistantrouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $ASSISTANTROUTER_API_KEY" \
  -d '{
    "model": "anthropic/claude-haiku-4.5",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "What is the capital of France?"}
    ]
  }'

In direct mode, RAG (file_search) and web search tools are not available. Use an assistant to access these features.

Response

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1704067200,
  "model": "anthropic/claude-haiku-4.5",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "The capital of France is Paris."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 25,
    "completion_tokens": 8,
    "total_tokens": 33
  },
  "x_assistantrouter": {
    "cost_cents": 0.15,
    "assistant_id": "your-assistant-uuid",
    "conversation_id": "conv_abc123",
    "tools_used": [],
    "was_fallback": false
  }
}

The x_assistantrouter extension includes:

cost_cents: The cost of this request in cents
assistant_id: The assistant ID used for this request
conversation_id: The conversation ID (if provided or auto-created)
tools_used: Array of tools that were invoked (e.g., ["web_search"])
was_fallback: Whether a fallback model was used

Streaming

Set stream: true to receive partial responses as Server-Sent Events (SSE):

const stream = await client.chat.completions.create({
  // @ts-ignore
  assistant_id: 'your-assistant-uuid',
  messages: [{ role: 'user', content: 'Tell me a story.' }],
  stream: true,
});

for await (const chunk of stream) {
  const content = chunk.choices[0]?.delta?.content || '';
  process.stdout.write(content);
}

stream = client.chat.completions.create(
    extra_body={"assistant_id": "your-assistant-uuid"},
    messages=[{"role": "user", "content": "Tell me a story."}],
    stream=True,
)

for chunk in stream:
    content = chunk.choices[0].delta.content or ""
    print(content, end="", flush=True)

Tool Calling

Enable function calling by providing tools in your request:

const response = await client.chat.completions.create({
  // @ts-ignore
  assistant_id: 'your-assistant-uuid',
  messages: [{ role: 'user', content: 'What is the weather in Paris?' }],
  tools: [
    {
      type: 'function',
      function: {
        name: 'get_weather',
        description: 'Get the current weather for a location',
        parameters: {
          type: 'object',
          properties: {
            location: {
              type: 'string',
              description: 'The city and country',
            },
          },
          required: ['location'],
        },
      },
    },
  ],
});

Errors

Code	Description
`authentication_error`	Invalid or missing API key
`insufficient_balance`	Not enough credits in wallet
`rate_limit_exceeded`	Request rate limit exceeded
`not_found`	Assistant not found
`bad_request`	Invalid request parameters

Chat Completions

Chat Completions

Request Body

Messages Format

Example Request (With Assistant)

Example Request (Direct Mode)

Response

Streaming

Tool Calling

Errors

On this page

Chat Completions

Chat Completions

Request Body

Messages Format

Example Request (With Assistant)

Example Request (Direct Mode)

Response

Streaming

Tool Calling

Errors

On this page