Qwen-Image-Edit on ModelScope
ModelScope API-Inference (modelscope.cn) Free API tier
The catch: Needs a verified email on your ModelScope profile, plus an Alibaba Cloud account bound to ModelScope with real-name (identity) verification. The daily points are credited only on days you log in and expire after 24 hours. ModelScope calls API-Inference a non-commercial, no-SLA service, and older models are removed as new ones arrive. Calls are asynchronous: the POST returns a task_id that you poll at /v1/tasks/{task_id} with header X-ModelScope-Task-Type: image_generation.
| Access | Free API tier |
|---|---|
| Free limits | Free with ModelScope points: 200 a day for logging in plus 50 a day once a real-name-verified Alibaba Cloud account is bound; a call costs 0.5, 1 or 2 points (light, mainstream or flagship model). ModelScope does not publish which tier this model is in on a public page. |
| Modality | image |
| Credit card | – |
| Commercial use | Not allowed |
| Trains on your data | – |
| Licence or terms | ModelScope User Agreement (modelscope.cn/docs/agreements/terms), plus each model's licence |
| Context window | – |
| Model id | Qwen/Qwen-Image-Edit |
| Base URL | https://api-inference.modelscope.cn/v1 |
| Last checked | 8 Oct 2026 |
How to use Qwen-Image-Edit
It speaks the OpenAI API at https://api-inference.modelscope.cn/v1, so the OpenAI SDK you already use works once you change the base URL and key.
curl https://api-inference.modelscope.cn/v1/images/generations -H "Authorization: Bearer $MODELSCOPE_TOKEN" -H "Content-Type: application/json" -H "X-ModelScope-Async-Mode: true" -d '{"model":"Qwen/Qwen-Image-Edit","prompt":"Make the sky purple","image_url":["https://resources.modelscope.cn/aigc/image_edit.png"]}'I want to use the free AI model "Qwen-Image-Edit" from ModelScope (model id: Qwen/Qwen-Image-Edit) in my project. Please: 1. Check the provider docs and give me the exact steps to get a free API key. 2. Give me a minimal working example that reads the key from an environment variable. 3. Tell me the real free limits and any catch (commercial use, card, data use). Listed limits: 250 points/day; 0.5–2 points per call. 4. If the same provider has a more generous free option, mention it.
Before you ship on it
- Needs a verified email on your ModelScope profile, plus an Alibaba Cloud account bound to ModelScope with real-name (identity) verification. The daily points are credited only on days you log in and expire after 24 hours. ModelScope calls API-Inference a non-commercial, no-SLA service, and older models are removed as new ones arrive. Calls are asynchronous: the POST returns a task_id that you poll at /v1/tasks/{task_id} with header X-ModelScope-Task-Type: image_generation.
- Limits page: API-Inference is itself a non-commercial, non-profit product; do not use it for online tasks that need high concurrency or an SLA; for commercial use, ModelScope suggests commercial platforms' APIs.
- Free tiers are shared capacity: treat the limits as a ceiling, not a promise, and keep a paid fallback.
Questions
Is Qwen-Image-Edit on ModelScope free?
Yes. ModelScope offers it as free api tier, with these limits: 250 points/day; 0.5–2 points per call.
Can I use Qwen-Image-Edit on ModelScope commercially?
No: the free access is for testing or non-commercial use. A paid plan is needed for production.
How do I start with Qwen-Image-Edit on ModelScope?
It speaks the OpenAI API at https://api-inference.modelscope.cn/v1, so the OpenAI SDK you already use works once you change the base URL and key.
Source: modelscope.cn/models/Qwen/Qwen-Image-Edit, checked 8 Oct 2026.