Well, look what we have here. Another day, another AI model to integrate into our coding workflow. Today, I'll show you how to set up Qwen 3 Coder Plus directly in Zed Editor. Because apparently, switching to a terminal to use AI is so 2023. Let's get this over with.

## What You'll Need

For this thrilling adventure, you'll need:
1. Qwen CLI installed (shocking, I know)
2. Zed Editor (because what else would you use?)

### Installing Qwen CLI

First, go get Qwen CLI from [here](https://github.com/QwenLM/qwen-code). Once it's installed, you'll need to log in by typing `/auth` in your terminal. Or maybe the system will hold your hand and redirect you automatically. Who knows? Technology is unpredictable like that.

> [!Note]
> **Fun fact:** The free version gives you 2000 daily requests with a limit of 60 requests per minute. How generous! It's almost like they want you to actually use their product.

### Installing Zed Editor

If you don't have Zed Editor yet, download it from [their website](https://zed.dev/download).

> [!Note]
> **Minor inconvenience:** Zed only supports macOS and Linux at the moment. Windows users, I guess you'll just have to sit this one out. Maybe next time?

## Now for the Main Event

Assuming you've successfully installed both pieces of software (congratulations, by the way), let's make them talk to each other.

### Step 1: Open the Agent Panel

In Zed Editor, find and open the "Agent Panel". It's probably hiding somewhere obvious, but you know how it is with UI design these days.


### Step 2: Navigate to Settings

Once the panel is open, click the "..." icon at the top and select "Settings" from the menu. Because every application needs its own special snowflake settings page, apparently.


### Step 3: Add a New Provider

Scroll down in the settings page until you see "LLM Providers" and click the "Add Provider" button. From the dropdown that appears, select "OpenAI". Yes, I know you're setting up Qwen, but just go with it. This is how software works in 2023.


### Step 4: Configure API Details

Now we get to the fun part - copying and pasting cryptic strings of text!

1. First, we need to get your API key from the `~/.qwen` folder on macOS. Navigate there and you'll see a few files:


2. To read the "oauth_creds.json" file, run `zed oauth_creds.json` in your terminal. You'll see something like this:


3. The `access_token` value in this file is your API key (I've conveniently hidden mine for security purposes, but yours will be there). You'll also need to copy the `resource_url` value.

4. Now, back to Zed. In the "API URL" field, enter `https://portal.qwen.ai` and add `/v1` at the end. So it should look like: `https://portal.qwen.ai/v1`

5. In the "API Key" field, paste the `access_token` you copied earlier.

6. For "Provider Name", you can enter whatever you want. "My Not-So-Secret AI Weapon" perhaps?

7. In the "Models" section:
* Set "Model Name" to `qwen3-coder-plus`
* Set "Max Completions Tokens" to `1000000`
* Set both "Max Output Tokens" and "Max Tokens" to `65536`
* Check the "Supports tools" and "Supports images" boxes at the bottom

8. Finally, click "Save Provider". If everything went according to plan, it should look something like this:


## And That's It!

Congratulations! You can now talk to Qwen 3 Coder Plus directly in Zed Editor. No more terminal switching for you. Your coding workflow is now approximately 0.5% more efficient. The future is here!

**Bonus tip:** You can add other Qwen models using the same method. Just change the model name and adjust the token values accordingly. Because variety is the spice of life, or so they say.

## If Things Go Wrong

If you run into any issues following these instructions, feel free to email me at `[e-mail]`. I'll try my best to help, but no promises. Technology is fickle, and so am I before my morning coffee.

## Final Thoughts

Thanks for reading this guide. I hope it was helpful, or at least mildly entertaining. Now go forth and code with your fancy new AI assistant. Try not to let it take over your job completely.

Happy coding!