API

Chat Completions

Send messages to your fine-tuned models via the chat completions endpoint.

Endpoint

POST /v1/chat/completions

Request format

from openai import OpenAI

client = OpenAI(
    base_url="https://app.commissioned.tech/v1",
    api_key="your-api-key",
)

response = client.chat.completions.create(
    model="your-model-id",
    messages=[
        {"role": "user", "content": "Write a summary of our Q4 results."}
    ],
)

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

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://app.commissioned.tech/v1",
  apiKey: "your-api-key",
});

const response = await client.chat.completions.create({
  model: "your-model-id",
  messages: [
    { role: "user", content: "Write a summary of our Q4 results." }
  ],
});

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

curl https://app.commissioned.tech/v1/chat/completions \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-model-id",
    "messages": [
      {"role": "user", "content": "Write a summary of our Q4 results."}
    ]
  }'

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	The ID of your fine-tuned model
`messages`	array	Yes	Array of message objects

Message format

Each message in the messages array:

Field	Type	Values	Description
`role`	string	`"system"`, `"user"`, `"assistant"`	Who sent this message
`content`	string	Any text	The message content

Response format

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Here's a summary of the Q4 results..."
      },
      "finish_reason": "stop"
    }
  ]
}

Multi-turn conversations

The API is stateless — it doesn't remember previous messages. To have a multi-turn conversation, include the full message history with each request:

messages = []

# Turn 1
messages.append({"role": "user", "content": "What were our top products last quarter?"})
response = client.chat.completions.create(model="your-model-id", messages=messages)

assistant_reply = response.choices[0].message.content
messages.append({"role": "assistant", "content": assistant_reply})

# Turn 2
messages.append({"role": "user", "content": "How did they compare to Q3?"})
response = client.chat.completions.create(model="your-model-id", messages=messages)

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

const messages: OpenAI.ChatCompletionMessageParam[] = [];

// Turn 1
messages.push({ role: "user", content: "What were our top products last quarter?" });
let response = await client.chat.completions.create({
  model: "your-model-id",
  messages,
});

messages.push({ role: "assistant", content: response.choices[0].message.content! });

// Turn 2
messages.push({ role: "user", content: "How did they compare to Q3?" });
response = await client.chat.completions.create({
  model: "your-model-id",
  messages,
});

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

Each request includes the full conversation. This means token usage grows with conversation length. For long conversations, consider summarizing earlier messages or trimming the history.

Finding your model ID

Get your model ID from:

Dashboard — shown on each model card
/v1/models endpoint — returns all your models with their IDs
Model card details — click into any model to see its full ID

Error responses

Status	Cause
`400`	Invalid request body (missing model or messages)
`401`	Missing or invalid API key
`404`	Model not found (check the model ID)
`429`	Rate limit exceeded
`500`	Server error (retry the request)

Authentication

Create and manage API keys for the Commissioned API.

Models Endpoint

List your fine-tuned models via the API.

Chat Completions

On this page