API overview

Base URL: https://mediamonster.app The in-app reference is mediamonster.app/developers, which opens these docs. Mint a key at mediamonster.app/keys.

API and MCP access comes with Minotaur, Kraken, and Behemoth. A Cyclops key is refused with 403 plan_required.

Authentication

Send the key as a bearer token. Keys start mm_live_ and are shown once.

Authorization: Bearer mm_live_…

See Authentication.

Endpoints

Method	Path	Key	What it does
`GET`	`/api/v1/skills`	No	Live skills, prices, option ids
`GET`	`/api/v1/me`	Yes	Plan and credit balance
`POST`	`/api/v1/convert`	Yes	Convert or resize one file. Free of credits
`POST`	`/api/v1/run/{skill}`	Yes	Run a skill. Charges credits
`POST`	`/api/v1/runs`	Yes	Run a line — saved recipe or inline steps
`GET`	`/api/v1/runs/{id}`	Yes	Poll a run and collect outputs
`GET`	`/api/v1/assets`	Yes	Vault contents and room left
`POST`	`/api/v1/assets`	Yes	Store a finished job in the Vault
`GET`	`/api/v1/jobs/{id}`	Yes	Poll a job and collect fresh download URLs
`GET`	`/api/v1/galleries`	Yes	Galleries and public URLs
`POST`	`/api/v1/galleries`	Yes	Create a gallery
`POST`	`/api/mcp`	Yes	MCP server, streamable HTTP

GET /api/v1/skills is the live list of option ids. It does not need a key.

Skills and prices

Same catalogue the Workbench and MCP read. Send option ids on POST /api/v1/run/{skill}. An unknown option is an error.

Slug	Takes	Credits	Options
`background-remover`	image	2 per file	output
`face-cutout`	image	2 per file	crop, who, output
`blur-background`	image	3 per file	strength, look
`smart-expand`	image	6 per file	shape, fill
`upscale`	image	25 per file	factor, mode
`alt-text`	image	1 per file	tone, rename
`restyle`	image	6 per file	style, strength
`background-swap`	image	6 per file	lighting
`new-perspectives`	image	6 per file	angle, framing
`object-remove`	image	6 per file	mode, shadows
`restore`	image	6 per file	job, grain
`colourise`	image	6 per file	look
`headshot`	image	6 per file	look, backdrop, attire
`try-on`	image	6 per file	garment, pose
`virtual-model`	image	6 per file	worn, scene
`poster`	image	6 per file	shape, style
`colouring-page`	image	6 per file	detail, border
`meme`	image	1 per file	size, case
`mockup`	image	6 per file	surface, scene
`swap`	image	6 per file	mode, match
`animate`	image	100 per file	length, motion, sound
`captions`	video or audio	8 per minute	output, style, motion, position
`transcript`	video or audio	4 per minute	format, speakers, timestamps
`watermark`	image	0 per file	position, size, opacity, colour
`extract-audio`	video	0 per file	format, quality, channels
`video-to-gif`	video	0 per file	output, length, fps, loop
`thumbnail`	video	3 per file	count, shape, prefer
`clean-audio`	video or audio	4 per minute	strength, extras, output
`trim-video`	video	0 per file	—
`split-video`	video	0 per minute	pieces, shape
`upscale-video`	video	150–1200 per minute	target, footage
`b-roll`	video	50 per clip + 5 per minute	count, look
`dub-video`	video	150 per minute	language
`vocal-remover`	video or audio	25 per minute	separate, format
`video-background-remover`	video	750 per minute	background, format
`slow-motion`	video	400–1600 per minute	speed
`lip-sync`	video	100–750 per minute	voice, quality
`remove-silences`	video or audio	5 per minute	pauses, fillers
`compress-video`	video	0 per file	target
`deinterlace`	video	2 per minute	mode, order
`stabilise-video`	video	3 per minute	strength
`smart-reframe`	video	5 per minute	shape, layout
`translate-captions`	video	10 per minute	language, output, look, motion, position
`talking-photo`	image	150–600 per file	length, voice
`image-qa`	image	1 per file	purpose, strictness
`auto-tag`	image	1 per file	depth, style
`classify`	image	1 per file	mode, fallback
`blur-faces`	image	2 per file	style, strength
`video-face-blur`	video	4 per minute	style, strength
`clips`	video	12 per clip + 3 per minute	count, length, shape, layout, cover
`youtube-thumbnail`	image or video	9 per thumbnail + 3 per minute	count, archetype, palette, subject
`video-multiplier`	video	32 per second + 20, per version	change, count
`graphic-multiplier`	image	12 per version	change, count
`podcast-cover`	image, video or audio	10 per file	look, size

See Skills for the same prices grouped by drawer.

Generate from a prompt

Send a prompt, and pick a model or leave it for the default — each model is billed at its own price. Image and video generation also take a file and start from it: a picture to work from, or the first frame of a clip. Voice and music take no file.

Slug	Takes	Credits	Options
`generate`	image	5–20 per file	model, shape, size
`text-to-video`	image	100–340 per file	model, length, shape
`text-to-voice`	none	4–25 per minute	model, voice, format
`music`	none	4–100 per minute	model, length, format

Model ids and per-model prices are on Generators.

How requests behave

Credits come off first

A skill is charged before the provider is called. A failed run is refunded. The response includes credits_charged.

Signed URLs last ten minutes

Download links expire after ten minutes. GET the job or run again for fresh ones.

Unknown options are errors

An option the skill does not have is refused. A typo must not quietly hand back the default look.

Unknown convert formats are errors

Image: png, jpeg, webp (default WebP). Video: mp4, webm (default MP4). Anything else is 400 unsupported_format.

Wait, or poll

A call waits about 150 seconds, then returns 202 with poll_url. Send wait=false to get a job or run id immediately.

Rate limits are per key

REST and MCP share the same per-key ceiling: 60 / 300 / 1,000 requests a minute on Minotaur / Kraken / Behemoth. 429 includes retry-after.

Agents should start at MCP.

Get started

Workbench

Skills

API

Account

Help

Authentication

Endpoints

Skills and prices

Generate from a prompt

How requests behave

​Authentication

​Endpoints

​Skills and prices

​Generate from a prompt

​How requests behave

Authentication

Endpoints

Skills and prices

Generate from a prompt

How requests behave