Midjourney · Image
Midjourney · Introduction
Midjourney image (v8.2 / v8.1 / v7 / niji 7): text-to-image, 4 images per call billed per call; aspectRatio accepted (default 1024x1024). Reference images are not supported yet.
How to call
- Submit:
POST /api/cqt/generator/mj, returns a taskId. - Query:
GET /api/cqt/info/mj?id=taskId, poll until status = succeeded.
Capability map
💰 See the pricing page for prices →
Common Errors
| Code | msg | Meaning | How to fix |
|---|---|---|---|
| 400 | model must be one of: [...] | model is not a sub-model of this channel | Use a model from the sub-model table |
| 400 | <model> does not support ... / does not accept ... | The sub-model does not support this mode / resolution / parameter (e.g. resolution sent to a default-output-only model) | Adjust per the table; the error lists the supported tiers |
| 400 | at most N imageUrls allowed / numImages must be between 1 and N | Too many reference images or images requested | Reduce per the table |
| 500 | generation_failed | Generation failed; data.errorMsg has details (credits auto-refunded) | Adjust based on the error and retry; contact support if it persists |
Errors are returned as a bizCode, e.g. bizCode=451, msg=prompt_unsafe.
Tip: after saving your API Key via Authorize, every capability page supports "Try it".