Get started

Quickstart

Send one call through Mubit and find its request and reply in the console.

Before you start. Run the example in a bash or zsh terminal on your computer or a server, not in a browser. You need Python 3.10+, Node.js 22+ or curl, and an inference key for your project. If you can't create keys or choose how calls are paid, ask an admin of your organization.

1. Set up in the console

Open app.mubit.ai and sign in. Complete the first-time setup if prompted.

On Get started, create a Mubit key. It gets the permission Inference: calls this project. Copy it when shown, then set it in the terminal where you will run the example:

export MUBIT_API_KEY='YOUR_MUBIT_KEY'

Under Choose how calls are paid, select Use the starter credit if it's offered, or Add your provider key. The console names the providers the starter credit covers; for any other provider, add its API key. See Provider-key setup.

2. Send one call

Choose your provider and language below. The example uses Mubit's base URL and your Mubit key; keep the model ID you want to call. Use the base URL your console's Connect page shows.

ℹ️Note

Mubit records request and reply content by default. Use synthetic data for this example, and keep secrets out of prompts.

On the starter credit, use a model it serves, such as gpt-6-luna. For other models or options, add your own OpenAI key. See On the starter credit and Connect with the OpenAI SDK.

Create a virtual environment and install the SDK:

python3 -m venv .venv && source .venv/bin/activate
pip install openai

Save this as mubit-quickstart.py:

mubit-quickstart.py

import os
from openai import OpenAI
 
client = OpenAI(
    base_url="https://api.mubit.ai/openai/v1",
    api_key=os.environ["MUBIT_API_KEY"],
)
 
instructions = (
    "Sort the support message into one category: "
    "billing, bug, account or other. Reply with the category only."
)
 
response = client.chat.completions.create(
    model="gpt-6-luna",
    messages=[
        {"role": "system", "content": instructions},
        {"role": "user", "content": "I was charged twice for my October invoice."},
    ],
)
print(response.choices[0].message.content)

Run it:

python mubit-quickstart.py

Install the SDK and tsx, which runs TypeScript files:

npm install openai tsx

Save this as mubit-quickstart.mts. The .mts ending runs it as an ES module, which its top-level await needs.

mubit-quickstart.mts

import OpenAI from "openai";
 
const apiKey = process.env.MUBIT_API_KEY;
if (!apiKey) throw new Error("MUBIT_API_KEY is not set");
 
const client = new OpenAI({
  baseURL: "https://api.mubit.ai/openai/v1",
  apiKey,
});
 
const instructions =
  "Sort the support message into one category: " +
  "billing, bug, account or other. Reply with the category only.";
 
const response = await client.chat.completions.create({
  model: "gpt-6-luna",
  messages: [
    { role: "system", content: instructions },
    { role: "user", content: "I was charged twice for my October invoice." },
  ],
});
console.log(response.choices[0].message.content);

Run it:

npx tsx mubit-quickstart.mts

No install needed. Run this in the terminal where you set MUBIT_API_KEY:

Terminal

curl https://api.mubit.ai/openai/v1/chat/completions \
  -H "Authorization: Bearer $MUBIT_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "gpt-6-luna",
    "messages": [
      {"role": "system", "content": "Sort the support message into one category: billing, bug, account or other. Reply with the category only."},
      {"role": "user", "content": "I was charged twice for my October invoice."}
    ]
  }'

On the starter credit, use a model it serves, such as claude-haiku-4-5. For other models or options, add your own Anthropic key. See On the starter credit and Connect with the Anthropic SDK.

Create a virtual environment and install the SDK:

python3 -m venv .venv && source .venv/bin/activate
pip install anthropic

Save this as mubit-quickstart.py:

mubit-quickstart.py

import os
from anthropic import Anthropic
 
client = Anthropic(
    base_url="https://api.mubit.ai/anthropic",  # no /v1
    api_key=os.environ["MUBIT_API_KEY"],
)
 
instructions = (
    "Sort the support message into one category: "
    "billing, bug, account or other. Reply with the category only."
)
 
message = client.messages.create(
    model="claude-haiku-4-5",
    max_tokens=64,
    system=instructions,
    messages=[
        {"role": "user", "content": "I was charged twice for my October invoice."},
    ],
)
print(message.content[0].text)

Run it:

python mubit-quickstart.py

Install the SDK and tsx, which runs TypeScript files:

npm install @anthropic-ai/sdk tsx

Save this as mubit-quickstart.mts. The .mts ending runs it as an ES module, which its top-level await needs.

mubit-quickstart.mts

import Anthropic from "@anthropic-ai/sdk";
 
const apiKey = process.env.MUBIT_API_KEY;
if (!apiKey) throw new Error("MUBIT_API_KEY is not set");
 
const client = new Anthropic({
  baseURL: "https://api.mubit.ai/anthropic", // no /v1
  apiKey,
});
 
const instructions =
  "Sort the support message into one category: " +
  "billing, bug, account or other. Reply with the category only.";
 
const message = await client.messages.create({
  model: "claude-haiku-4-5",
  max_tokens: 64,
  system: instructions,
  messages: [
    { role: "user", content: "I was charged twice for my October invoice." },
  ],
});
for (const block of message.content) {
  if (block.type === "text") console.log(block.text);
}

Run it:

npx tsx mubit-quickstart.mts

No install needed. Run this in the terminal where you set MUBIT_API_KEY:

Terminal

curl https://api.mubit.ai/anthropic/v1/messages \
  -H "x-api-key: $MUBIT_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-haiku-4-5",
    "max_tokens": 64,
    "system": "Sort the support message into one category: billing, bug, account or other. Reply with the category only.",
    "messages": [
      {"role": "user", "content": "I was charged twice for my October invoice."}
    ]
  }'

If the console doesn't list Gemini under the starter credit, add your Gemini key under Provider keys first. On the starter credit, use a model it serves, such as gemini-3.8-flash. See On the starter credit and Connect with Gemini.

Create a virtual environment and install the SDK:

python3 -m venv .venv && source .venv/bin/activate
pip install google-genai

Save this as mubit-quickstart.py:

mubit-quickstart.py

import os
from google import genai
from google.genai import types
 
client = genai.Client(
    api_key=os.environ["MUBIT_API_KEY"],
    http_options=types.HttpOptions(
        base_url="https://api.mubit.ai/gemini",  # no /v1beta
    ),
)
 
instructions = (
    "Sort the support message into one category: "
    "billing, bug, account or other. Reply with the category only."
)
 
response = client.models.generate_content(
    model="gemini-3.8-flash",
    contents="I was charged twice for my October invoice.",
    config=types.GenerateContentConfig(system_instruction=instructions),
)
print(response.text)

Run it:

python mubit-quickstart.py

Install the SDK and tsx, which runs TypeScript files:

npm install @google/genai tsx

Save this as mubit-quickstart.mts. The .mts ending runs it as an ES module, which its top-level await needs.

mubit-quickstart.mts

import { GoogleGenAI } from "@google/genai";
 
const apiKey = process.env.MUBIT_API_KEY;
if (!apiKey) throw new Error("MUBIT_API_KEY is not set");
 
const ai = new GoogleGenAI({
  apiKey,
  httpOptions: { baseUrl: "https://api.mubit.ai/gemini" }, // no /v1beta
});
 
const instructions =
  "Sort the support message into one category: " +
  "billing, bug, account or other. Reply with the category only.";
 
const response = await ai.models.generateContent({
  model: "gemini-3.8-flash",
  contents: "I was charged twice for my October invoice.",
  config: { systemInstruction: instructions },
});
console.log(response.text);

Run it:

npx tsx mubit-quickstart.mts

No install needed. Run this in the terminal where you set MUBIT_API_KEY:

Terminal

curl "https://api.mubit.ai/gemini/v1beta/models/gemini-3.8-flash:generateContent" \
  -H "x-goog-api-key: $MUBIT_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "system_instruction": {"parts": [{"text": "Sort the support message into one category: billing, bug, account or other. Reply with the category only."}]},
    "contents": [{"parts": [{"text": "I was charged twice for my October invoice."}]}]
  }'

3. Check the result

For this support message, a successful reply should classify it as billing; exact output can vary.

In the console, open Requests and select your newest call. Check its Request, Reply and Served by details. Only calls made through the configured client go through Mubit.

If the call fails, use its error code to find the fix.

Create your first adapter

Send examples of one task, follow its training, and compare a passing version with your provider model.

1. Set up in the console

2. Send one call

3. Check the result

Next