Quickstart

Create a key, configure your environment, and run your first request in minutes.

1. Create an API key

In myt AI Cloud, open API keys, create a key for your project, and store it securely. Keys are scoped to your organisation and project; you will use this as MYT_API_KEY below.

2. Install an SDK

You can use the official OpenAI SDKs pointed at the myt AI Cloud base URL.

npm install openai

pip install openai

3. Configure environment variables

Enter your API key and base URL — code examples on this page update as you type:

API key

Base URL

Full inference endpoint URL, including /v1 (same as on API Keys).

Values are saved in your browser and update the code examples below.

export MYT_API_KEY="sk-..."
export MYT_BASE_URL="https://api.example.com/v1"

MYT_BASE_URL is the root for OpenAI SDK clients. Code examples use a model id from your catalog; swap it for any model your project can access (see Models).

4. First API request

examples/first-request

curl -sS --request "POST" \
  --url "$MYT_BASE_URL/chat/completions" \
  --header "Authorization: Bearer $MYT_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{"model":"myt/qwen3.6-35b-a3b","messages":[{"role":"user","content":"Say hello in one sentence."}]}'

5. Streaming example

Stream tokens as they arrive:

examples/streaming

curl -sS --request "POST" \
  --url "$MYT_BASE_URL/chat/completions" \
  --header "Authorization: Bearer $MYT_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{"model":"myt/qwen3.6-35b-a3b","stream":true,"messages":[{"role":"user","content":"Count from 1 to 5 slowly."}]}'

6. Next steps

Read Authentication for bearer tokens and headers
Browse the Models catalog and pick a default for your workload
Jump to Chat completions for full request and response fields