23 studios · 265+ models · 20+ providers
Twenty-three studios.
One generative workspace.
Video, image, audio and 3D — plus editing, lip sync and automation. 265+ models from 20+ providers behind a single interface, so you switch models instead of switching tools.
100 free credits on signup. No card required.
- studios
- 23studios
- models
- 265+models
- voices
- 472voices
- languages
- 39languages
- workflow templates
- 48workflow templates
Every studio
23 studios, one workspace
Each studio is built for one job and renders its controls from the selected model’s own schema — so switching models changes the form rather than silently dropping what you set.
Generate
10 studiosMake something from a prompt, a still, a clip or a script.
Video
Text, image and video to video
Image
Text to image and image editing
Audio
Music, sound effects and speech
3D
Text or image to 3D model
Cinema
Direct the virtual camera
Vibe Motion
Drive a still with an effect
Marketing
Ad creative in every format
Seedance
Seedance 2 — text, image and reference to video
UGC Ads
Vertical ad video from an actor still and a script
Voiceover
Script to speech across 472 voices
Edit
6 studiosFix, upscale, retime and restage work you already have.
Automate
7 studiosChain models, brief an agent, or drive the API directly.
Prompt Architect
Refine an idea into a structured, reusable prompt
GEO Audit
Score a page for AI-search visibility
Workflows
Run multi-step MuAPI workflows
Agents
Browse, clone and chat with skilled MuAPI agents
Design Agent
Autonomous agent that designs on a board
Raw API
Hand-written JSON to a MuAPI endpoint
Pricing
Live per-model cost for the whole catalog
The catalog
The models you actually want, already wired up
265+ models from 20+ providers, kept current rather than curated once. No model is behind a separate subscription.
- Google Veo
- OpenAI Sora
- Kling
- Seedance
- Midjourney
- Flux
- Hailuo
- Wan
- Runway
- Seedream
- Imagen
- Ideogram
- Leonardo
- PixVerse
- Vidu
- Grok
- Stability
- Qwen
- and more
Switch models, not tools
Iterate on a cheap standard-tier model, then re-run the shot you like on a premium one. Same studio, same form, one dropdown.
Controls come from the schema
Each model publishes its own inputs, and the form is generated from them — so there are no controls that look adjustable but get discarded before the request is sent.
Live per-model pricing
Cost is read from the provider rather than a hardcoded table, and a per-model estimate prices a specific parameter set before you submit.
Voiceover
472 voices. 39 languages. One script.
Paste a script, pick a speaker, get broadcast-ready speech. These are real language buckets, not one English voice with an accent applied — and each voice takes an emotion, so the same line can be read seven different ways.
- happy
- sad
- angry
- fearful
- disgusted
- surprised
- neutral
- Portuguese
- English
- Spanish
- Korean
- Mandarin
- Japanese
- Indonesian
- Cantonese
- Russian
- French
- Italian
- Thai
- Hindi
- German
- Polish
- +24 more
How it works
Three steps, then it runs without you
The durability matters more than it sounds. Video generation takes minutes, and a job that dies when the tab closes is a job you have to babysit.
Pick a studio
23 of them, grouped by what you are doing: generating, editing, or automating. The studio decides which models are offered.
Describe it, or upload it
Write a prompt, drop in a reference image, or supply a clip. The form is generated from the selected model, so you only see parameters that model actually accepts.
Close the tab if you want
Generation runs server-side and durably — each step is checkpointed, so a refresh, a dropped connection or a closed laptop does not lose the job. Finished work lands in your gallery.
FAQ
Questions, answered plainly
What do credits cost?
Every account starts with 100 free credits. Bloomtle does not currently sell credits — there is no checkout, and nothing on this site asks for a card.
How far do the free credits go?
Roughly fifty images at 2 credits each, twenty-five voiceovers at 4, or three premium 5-second video clips at around 30. Prompt refinement is the cheapest thing here at 3 credits.
Do I need my own API keys?
No. Every model runs on a platform key, so there is nothing to configure and no separate provider account to hold.
Which models are included?
All 265+ of them, across 20+ providers. No model sits behind a higher tier — premium families simply cost more credits per generation.
What happens if a generation fails?
The credits are refunded automatically. A job that fails mid-run is marked failed and its debit reversed, and a scheduled sweep catches anything left stranded.
Do I have to keep the tab open while it generates?
No. Jobs run server-side and are checkpointed step by step, so a refresh, a closed tab or a dropped connection does not lose the work.
Is there an API?
The Raw API studio posts a hand-written JSON body to an allowlisted model endpoint, for cases where the generated form does not expose a parameter you need.
Start generating today
100 free credits on signup — roughly fifty images, or twenty-five voiceovers, or three premium video clips.
No card required.