Access Midjourney V8.2 - also searched as Midjourney 8.2 or MJ V8.2 - through EvoLink's unified image API. Test brand mood visuals, concept art and style-consistent illustration sets before you integrate; variation, remix, canvas edit, retexture, upload paint and remove background share the same key.
Pick a model ID, explore composition in Draft, then finalize in HD; a finished task can go straight into variation, remix or canvas edit — all with live credit estimates.
Create an image task
Fixed to mj-v8.2Text-to-image and image-to-image. 4 images per request, 24 sketches in Draft.
Apart from Speed and Quality, every Midjourney setting lives in the prompt as a --parameter. The controls below only write into the prompt; you can also type them by hand.
58 / 1024 description characters
--ar--qFree — higher values only add render time; 4 is the high-quality mode.
Up to 20 public image URLs placed before the text. One image needs a text description.
Results are kept for 24 hours — download anything you want to keep.
Every model ID is billed per request, not per image. Generation multiplies a base rate by speed (Draft price ×1, Fast price ×1) and quality (Standard price ×1, HD price ×1.5); the --q detail parameter (1–4) is free. Editing routes are flat rate.
First Test Cost
mj-v8.2 · Fast · Standard · 4 imagesOne Fast Standard request (4 images)
About 125 tests with $10 in credits.
In Playground, keep Image Generation selected with Fast speed and Standard quality. Draft costs the same and returns 24 small sketches for exploring composition; switch to HD only for the final render. --q 2–4 in the prompt adds render time but no cost.
Midjourney V8.2 testing budget guide
Pick an amount based on how many tests you expect.Good for a first validation.
Good for prompt iteration.
Good before production integration.
| Item | Rule | Rate | Billing |
|---|---|---|---|
Draft · Standardmj-v8.2 | Draft · 24 sketches at 0.5K, same price as Fast; not combinable with HD. | $0.080/request-10% 5.4 cr/request$0.089 official price | Per request |
Fast · Standardmj-v8.2 | Fast · Standard — the default tier. | $0.080/request-10% 5.4 cr/request$0.089 official price | Per request |
Fast · HDmj-v8.2 | Fast · HD — native 2K output, price ×1.5. | $0.120/request-10% 8.1 cr/request$0.133 official price | Per request |
mj-v8.2mj-v8.2mj-v8.2| Item | Rule | Rate | Billing |
|---|---|---|---|
Variationmj-v8.2-variation | Subtle or strong variation of a finished image; flat rate. | $0.120/request-10% 8.1 cr/request$0.133 official price | Per request |
Remix · Fastmj-v8.2-remix | New prompt on a finished image; flat rate (Fast). | $0.120/request-10% 8.1 cr/request$0.133 official price | Per request |
Canvas Edit · Fastmj-v8.2-edit | Canvas edit with img_pos and optional mask; flat rate (Fast). | $0.120/request-10% 8.1 cr/request$0.133 official price | Per request |
Retexture · Fastmj-v8.2-retexture | Any image URL, new materials and style; flat rate (Fast). | $0.120/request-10% 8.1 cr/request$0.133 official price | Per request |
Upload Paint · Fastmj-v8.2-upload-paint | Uploaded image with a required mask; flat rate (Fast). | $0.120/request-10% 8.1 cr/request$0.133 official price | Per request |
Remove Backgroundmj-v8.2-remove-bg | One image URL, no prompt, flat rate. | $0.080/request-10% 5.4 cr/request$0.089 official price | Per request |
mj-v8.2-variationmj-v8.2-remixmj-v8.2-editmj-v8.2-retexturemj-v8.2-upload-paintmj-v8.2-remove-bgMidjourney V8.2 leads with aesthetics: bolder, more sophisticated defaults, --q detail levels 1–4 and native HD output. It shines when the image is the deliverable — brand mood visuals, concept design and illustration sets that stay on one style — with Style Reference, Draft exploration and an editing chain on the same route.
mj-v8.2Text-to-image and image-to-image with native prompt syntax. 4 images per request, or 24 sketches in Draft. Speed and quality are API parameters; detail is the --q prompt parameter.
mj-v8.2-variationSubtle or strong variations of one image from a completed mj-v8.2 task. The prompt is ignored; flat rate.
mj-v8.2-remixReinterpret one image from a completed task with a new prompt while keeping its composition.
mj-v8.2-editPlace a finished image on a canvas with img_pos, optionally mask a region, and fill the rest from the prompt.
mj-v8.2-retextureSend any public image URL and a prompt to change materials and style while preserving structure.
mj-v8.2-upload-paintUpload an image, define a mask and canvas position, and repaint the masked region from the prompt.
mj-v8.2-remove-bgCut out the subject of one public image URL and return a transparent PNG. No prompt, flat rate.
These fields cover the main integration decisions. Open the API tab for the full request and response schema of each model ID.
modelstringmj-v8.2 for generation; -variation, -remix, -edit, -retexture, -upload-paint and -remove-bg for derived routes.
promptstringNative Midjourney syntax such as --ar 16:9 --s 500 --sref URL. Image URLs go at the start for image-to-image. Up to 1024 characters of description.
model_params.speedenumdraft (24 sketches, price ×1) or fast (price ×1). Editing routes accept fast only; variation and remove-bg have no speed.
qualityenumstandard (price ×1) or hd for native 2K (price ×1.5). Generation only; hd cannot be combined with draft.
--qprompt parameter1–4 written in the prompt (e.g. --q 3), maps to native --quality. Richer detail and longer render; never changes the price. V8.1 accepts only 1 or 4.
model_params.task_idstringA completed mj-v8.2 task you own, plus image_number (0–3 for a Fast task, 0–23 for a Draft task). Required by variation, remix and canvas edit.
image_urlsarrayOne public http(s) image URL for retexture, upload paint and remove background. Top-level field, not inside model_params.
callback_urlstringHTTPS webhook that receives the task result instead of polling.
Use the EvoLink API for product integration and batch jobs, or call it from Codex, Claude, or Gemini for fast creative and development workflows. Both paths share the same EvoLink API key, balance, model routes, and task history.
Best for: product backends, batch jobs, automated workflows
Call EvoLink’s unified image API from your server and control the model ID, parameters, task queue, callbacks, and result storage.
Best for: creative and development tasks in Codex, Claude, and Gemini
Give the Agent your output goal, assets, and acceptance criteria. It can choose the route, assemble the request, track the task, and return the result without requiring you to hand-code every step.
This example shows the shortest runnable flow: create a task, save the returned task ID, then poll or wait for an HTTPS callback. Open the API tab for the complete parameter and response reference.
curl -X POST https://api.evolink.ai/v1/images/generations \
-H "Authorization: Bearer $EVOLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "mj-v8.2",
"prompt": "A cinematic shot of a Maine Coon cat on a neon-lit balcony --ar 16:9 --s 500",
"quality": "standard",
"model_params": {
"speed": "fast"
},
"callback_url": "https://your-domain.com/webhooks/image-task-completed"
}'
# Save the returned task ID, then query:
curl https://api.evolink.ai/v1/tasks/{task_id} \
-H "Authorization: Bearer $EVOLINK_API_KEY"Check every value against the parameter cards above: ranges, allowed options, asset counts, and file types must match the selected route.
Check the Authorization bearer token and confirm the available balance and task billing in the console.
Review likeness rights, brand assets, readable text, sensitive content, and input-asset compliance.
A Midjourney V8.2 task that reaches a final failed state is not charged. Keep the task ID, inspect the final state, and identify whether the cause is validation, review, or execution before retrying.
callback_url must be a public HTTPS URL, not localhost or a private network address. Always keep the task ID as a polling fallback.
| Feature | Midjourney V8.2 | Midjourney V8.1 | Seedream 5.0 Pro |
|---|---|---|---|
| Starting price | ~$0.080 / request (4 images) | Same base rate | Per image, 1K / 2K tiers |
| Speed tiers | Draft / Fast | Draft / Fast | Single tier |
| Detail level (--q) | 1–4, free | 1 or 4 only | Quality tiers |
| References | Image prompts + --sref | Image prompts, --sref, --oref | Up to 10 input images |
| Best for | Aesthetic-first creative work, illustration series | Existing V8.1 pipelines | Product creative, instruction edits |
Draft and Fast are price ×1; Standard is price ×1, HD is price ×1.5. Draft cannot be combined with HD.
Variation, remix and canvas edit of a V8.2 image are rendered by the V7 engine upstream (documented by the upstream channel). Style may differ slightly; pricing is unchanged.
--oref, --cref, --stop and other unsupported parameters are sent upstream and rejected with the upstream message; the task fails and reserved credits are refunded. Version, speed and quality parameters are stripped and controlled by API parameters; --q is passed through.
An official Midjourney route with domestic reachability — no subscription-account pools, no terms-of-service risk for your product.
Generation, variation, remix, canvas edit, retexture, upload paint and remove background on one endpoint and one key.
Reserve, settle, and refund in full when a task fails — including upstream parameter rejections.
Poll or receive webhooks for pending, processing, completed and failed; results are signed URLs valid for 30 days, so save them promptly.
Draft for exploration at the base rate, HD only when the render matters; the --q detail parameter is free.
Compare Midjourney generations under one EvoLink account and API key, then choose by speed tiers, reference controls and editing routes.
Midjourney V8.2
CurrentMidjourney V8.2 leads with aesthetics: bolder, more sophisticated defaults, --q detail levels 1–4 and native HD output. It shines when the image is the deliverable — brand mood visuals, concept design and illustration sets that stay on one style — with Style Reference, Draft exploration and an editing chain on the same route.
Midjourney V8.1
Previous generation with Draft / Fast speed, HD output and Omni Reference (--oref) support.
View API
Midjourney V7
Midjourney V7 route with twelve operations including upscale, pan, outpaint and inpaint.
View API
Seedream 5.0 Pro
BytePlus image route with 1K/2K output tiers, up to 10 reference images and instruction-based editing.
View API
Nano Banana Pro
Premium Nano Banana route for higher-fidelity images and production-grade output.
View API
Qwen Image Edit
Image editing route for direct transformations and localized creative changes.
View APIV8.2 improves aesthetics and image quality, accepts --q 1–4 in the prompt (V8.1 only 1 or 4) at no extra cost. Base prices are the same as V8.1. Omni Reference (--oref) and Character Reference (--cref) are not supported on V8.2 and are rejected by the upstream.
--p is passed through unchanged, but it only accepts a moodboard ID created by the upstream channel under this platform's organization account. The platform does not yet expose an endpoint for creating moodboards, and personalization codes from midjourney.com do not apply. For a consistent look across a series, use --sref / --sw and --seed.
Billing is per request. Generation starts at the Fast Standard rate for 4 images; HD is price ×1.5. Draft costs the same as Fast and returns 24 sketches. Variation, remove-background, remix, canvas edit, retexture and upload paint are all flat per request.
No. --q 1–4 only changes render time and richness. Write it in the prompt on mj-v8.2 generation; derived routes inherit the source image.
No. Draft returns 24 lightweight 0.5K sketches and cannot be combined with hd quality; the request is rejected before any credits are reserved.
mj-v8.2-variation, mj-v8.2-remix and mj-v8.2-edit take model_params.task_id of a completed mj-v8.2 task you own plus image_number (0–3 for a Fast task, 0–23 for a Draft task). Retexture, upload paint and remove background take a public image URL in image_urls instead.
Derived tasks based on V8.2 images (variation, remix, canvas edit) are rendered by the V7 engine upstream — this is documented by the upstream channel. Pricing is unchanged.
--ar, --s, --chaos, --exp, --raw, --seed, --iw, --sref, --sw, --tile, --weird and --q work inside the prompt. Speed and quality are API parameters and their parameters are stripped. Unsupported parameters such as --oref, --cref and --stop are passed through and rejected by the upstream with a clear message; the task fails and credits are refunded.
Fast returns 4 images; Draft returns 24 sketches. Some images may be filtered by Midjourney's content review, so you may get fewer back; billing is per request.
Reserved credits are refunded in full for failed tasks, including upstream parameter rejections and timeouts. Requests where content review filters some images are still billed.
Return to the Playground for more tests, or create an API key and call /v1/images/generations with model mj-v8.2.