Skip to content

MiniMax H3 Max Uncensored

MiniMax's pricier tier: generates from a prompt or animates a reference, 5-15s, up to 4K.

VideoText to video · Image to videoMiniMaxfrom 10 credits / s
vellria-minimax-h3-max

MiniMax H3 Max is developed by MiniMax. Vellria provides access to the model; it does not own it. All trademarks belong to their respective owners.

Input

Form

0 / 10000

Reference

0–1 images. You can upload files or paste https URLs; the provider downloads the image from the URL.

Settings

No generation yet

No generation yet

Output

No generation yet

Cover art

from 10 credits / s

Sign in to run

Your request will cost 80 credits.

Overview

This is the MiniMax line at its dearer setting, so the working pattern is to settle the shot on the cheaper one and spend here on the take you mean to keep. The two lines share a resolution ladder rung for rung, which is what makes that possible: a framing found lower down comes back at the same size here and only the bill changes. Sound is not a setting in this family — it arrives with every clip — so write the action knowing it will be heard as well as seen.

Prompt or reference

With only a description, a run starts from your wording and the frame shape is a choice you make there. Attach a picture and it becomes the opening frame instead, fixing the proportion with it, so there is nothing left to choose and a crop belongs before the upload rather than after the render.

Specifications

  • Resolution: 480p768p1440p4K
  • Duration: 5–15s
  • Aspect ratio (Text to video): 21:916:94:31:13:49:16
  • Reference images: 1–1 reference images

How the video side of the catalog works as a whole, and what it allows, is on the uncensored AI video generator page; explicit clips have their own page on NSFW video generation, and starting from a still is covered under NSFW image to video.

Pricing

ResolutionCredits
480p10 credits / s
768p16 credits / s
1440p30 credits / s
4K41 credits / s

Call it from the API

Send vellria-minimax-h3-max as the model field to POST /v1/generate/video, with your API key as a Bearer token. The request below uses this model's default settings from the catalog and returns a generation id to poll.

curl https://vellria.com/v1/generate/video \
  -H "Authorization: Bearer vll_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "vellria-minimax-h3-max",
    "prompt": "…",
    "duration": 5,
    "resolution": "768P",
    "ratio": "16:9"
  }'

Every parameter, the polling call and the upload endpoint are in the API documentation; every video model id with its price is on the NSFW video API page.