Qwen API Key Setup

Complete guide to creating your API key in Alibaba Cloud Model Studio, the cheapest provider in CheesCoach.

$ The cheapest option in all of CheesCoach: Qwen 3.7 Flash costs $0.03/M input, $0.13/M output
💰 New accounts include free tokens to try the models
🔗 API compatible with OpenAI format
Part 1

Creating your API Key

1
Create an Alibaba Cloud account

Visit Alibaba Cloud Model Studio and sign up with your email. Model Studio is the console that serves the Qwen models.

2
Choose the Singapore region

The first time you enter, the console asks you to pick a region. Stay on Singapore (ap-southeast-1): that is the international endpoint CheesCoach talks to. A key created in another region will not work.

3
If the console shows up in Chinese

It happens, and it is disconcerting the first time. Use the language selector at the top of the page to switch to English. The button you are looking for reads "Create API Key" in English and "创建 API Key" in Chinese.

4
Open the API Keys page

Go to the API Key section of the console (left sidebar, "API Key").

5
Create API Key

Click "Create API Key". Keep the default workspace, write any description you like (e.g. "CheesCoach") and leave the permissions as "All".

6
Copy the key NOW

The key is shown only once. Copy it or download it right there, before closing the dialog. If you lose it there is no way to see it again — you can only delete it and create a new one.

7
Paste in CheesCoach

Open CheesCoach → Settings → AI section → Select "Qwen 3.7 Flash" → Configure → Paste your API key.

⚠️
The key is shown only once

This is the step that causes the most trouble. Alibaba Cloud does not let you look at an existing key again: copy it or download it the moment it appears on screen.

Part 2

Choosing your model

A
Qwen 3.7 Flash
Recommended
The cheapest model in the whole app: $0.03 per million input tokens and $0.13 per million output tokens. It answers fast and it is more than enough for move-by-move coaching.
CheesCoach turns Qwen's reasoning mode off on this model, so each analysis stays as small (and as cheap) as it looks.
B
Qwen 3.7 Plus
The powerful tier: $0.32 per million input tokens and $1.28 per million output tokens. It thinks before answering, so explanations are deeper — and each analysis costs noticeably more than with Flash.
Both models use the same API key. You can switch between them in CheesCoach settings at any time.

Pricing

Model Input per 1M tokens Output per 1M tokens
Qwen 3.7 Flash $0.03 $0.13
Qwen 3.7 Plus $0.32 $1.28
💰 Qwen 3.7 Flash is the cheapest option in CheesCoach — several times cheaper than DeepSeek, which used to hold that spot.

Important notes

⚠️ Copy the key when it is created: Alibaba Cloud never shows it again.
🌏 Keep the Singapore (ap-southeast-1) region: it is the international endpoint CheesCoach uses.
💬 The console sometimes opens in Chinese. Switch the language at the top of the page and carry on.
🚀 The free tokens that come with a new account are enough for plenty of testing.