What to do when Qwen Pro is sold out and you still need a working route today.
If you liked the Qwen Code workflow but the paid route is out of stock, the practical question is not “which vendor won?” It is whether you need the exact official plan or simply one OpenAI-compatible coding endpoint you can buy and test today in Qwen Code, LiteLLM, or a similar client.
Wait for official availability and accept their billing shape, limits and stock timing.
Keep the IDE, but point it at another OpenAI-compatible endpoint you can configure yourself.
Start with the smallest possible prepaid route before treating this like a larger subscription decision.
Need a route right now? If the official plan is sold out, pay the $5 starter amount first, then run one small request in Qwen Code. If the endpoint is useful, keep going. If not, stop after the first test.
1. Keep the IDE, change the endpoint
Qwen Code already supports OpenAI-compatible custom providers. If the blocker is plan availability or minimum spend, you can preserve the Qwen Code workflow while switching the endpoint underneath it.
{
"$version": 3,
"env": {
"BLACK_EAGLE_API_KEY": "sk-your-token"
},
"security": {
"auth": {
"selectedType": "openai"
}
},
"modelProviders": {
"openai": [
{
"id": "deepseek-v4-flash-black-eagle",
"name": "Black Eagle DeepSeek Flash",
"envKey": "BLACK_EAGLE_API_KEY",
"baseUrl": "https://api.blackeaglecambodia.com/v1"
}
]
}
}2. Choose by blocker
| If your blocker is... | Best next move | What you give up |
|---|---|---|
| Official Qwen plan is unavailable or sold out | Switch Qwen Code to another OpenAI-compatible endpoint. | You are no longer testing the exact official Qwen hosted path. |
| Minimum spend is too high for light personal use | Start with a small prepaid route and validate your workflow first. | Some model families may expand by request instead of instant self-serve. |
| You need the exact Qwen model family for a real benchmark | Use direct provider or request explicit Qwen route support. | Less flexibility on price and availability. |
3. What Black Eagle AI is and is not
Black Eagle AI is most useful when you want one small OpenAI-compatible paid test without turning Chinese-model billing and account setup into the whole project. DeepSeek text routes are live now. Qwen, Kimi, GLM and other families can be requested as demand becomes concrete.
4. Start with one real request
If your goal is to keep coding inside Qwen Code, you do not need to commit to a large plan first. Start by verifying one real hosted endpoint works inside the IDE.
curl https://api.blackeaglecambodia.com/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer sk-your-token" \
-d '{
"model": "deepseek-v4-flash",
"messages": [{"role": "user", "content": "Write a short Python CLI parser."}]
}'Concrete budget signal: at current public rates, a $5 starter test covers about 3.8M input + 3.8M output tokens on Economy Text or about 1.5M + 1.5M on Reasoning Text. That is usually enough to prove whether the Qwen Code workflow is worth keeping.