Create beyond limits.
AI image, video, audio and music generation in one studio
Write a prompt, choose the model that fits the task and check the credits before generating. Images, video, audio and music from leading AI models are available in one place, with one history.
- image models
- 18
- video models
- 35
- audio models
- 2
Made with Arumi
Results from the image, video, audio and music generators, shown with the prompts that produced them. Select an item to view it at full size.
AI generators for images, video, audio and music
Create images, videos, voice-overs and songs from a text prompt in one place, with the credit cost shown before each generation.
AI Image Generator
Text to image and AI photo editing. Describe a picture to generate it, or add reference images to edit a photo or keep a subject consistent.
Open the AI Image GeneratorAI Video Generator
Text to video and image to video. Turn a prompt or a still image into a video clip, guided by reference images, video and audio.
Open the AI Video GeneratorAI Audio Generator
Text to speech for voice-overs, narration and read-alouds. Paste a script and download it as natural-sounding speech.
Open the AI Audio GeneratorAI Music Generator
AI songs and instrumentals. Describe the mood and style to create music, or write your own lyrics for a song with vocals.
Open the AI Music Generator
Ready-to-use tools
Each tool performs a single task and is ready to use as soon as it opens.
- Background removerCut out the subject and export it as a transparent PNG at 1K.
- Image upscalerMake a small or blurry image sharp, up to 4K, without changing what's in it.
- Object removerRemoves the objects you describe and fills the area to match its surroundings.
- Background changerPlace a product or a person in a new scene, with matching light and shadows.
- Old photo restorationRepair scratches, tears and fading, and add natural colour to old photos.
- Photo to animeTurns a photo into anime, sketch, watercolour, 3D cartoon, comic or oil painting.
- Photo to collage posterContinues a photo into a 3:4 poster: the photo above, a paper collage below.
- Video upscalerBring a video up to 1080p, 2K or 4K with AI for a sharper picture.
- Animate old photosAnimate an old photo. The moment in the photo continues with natural movement.
- AI freestyle videoCreate a street freestyle rap video from one photo of yourself.
- Split imageSplit one image into a grid of up to 64 tiles and export the tiles you select.
- Join imagesStitch several images into one, horizontally or vertically.
- Add watermarkAdd a text or logo watermark to up to 20 images at once, as a single mark or tiled. Runs in your browser; the images never leave your computer.
- Audio trimTrim an audio file and export WAV, MP3, M4A, AAC, OGG, FLAC or WebM.
- MP3 to VideoCombine audio with one or more images to make an MP4 for video sites. Runs in your browser; the file never leaves your computer.
Features of the studio
Leading models side by side
Each generator brings together the major models for its medium: currently 55 models from 8 companies. Switching models keeps the prompt, and each model shows only the settings it supports.
Price shown before generating
The Generate button shows how many credits a run costs before it is clicked. If a setting is left to the model, the quote uses the most expensive option the model could choose.
The quoted price is charged
A successful generation is charged exactly the quoted credits. Every charge is listed on the Usage & billing page.
Generation in the background
Generations run on our servers. It is possible to switch pages, start the next take or close the tab. Finished results are available in Recent and in History.
Private uploads
Uploads are stored privately and open only to you while you are signed in. We do not train models of our own on prompts, uploads or results.
Workflows and free tools
A workflow shared in Explore can be copied to the canvas and run there. Free browser tools split, join and trim files without uploading them.
How to create with Arumi
Choose what to create
Select image, video, audio or music at the top of this page, or open a generator from the menu. Each generator lists its models and what each does well.
Write a prompt or add references
Describe the intended result. Models that accept reference images, video or audio provide upload slots, and the slots show only what the selected model can use.
Check the credits and generate
The price updates when the model, length or resolution changes. Select Generate. Work can continue while the generation renders.
Download, reuse or refine
Results appear in Recent and in History, where they can be downloaded, renamed, or have their settings reused for the next take.
Frequently asked questions
Need more information? Email us at support@arumi.ai.
What is Arumi?
Arumi is a web studio for creating images, video, audio and music with AI models built by different companies, under one account, one credit balance and one history. It also includes an AI canvas for building workflows, Explore for the creations and workflows that people share, and free tools that run in the browser.
Do I need an account?
The free tools work without one. Generating with AI models and running workflows requires signing in with Google or an email link, so that results and credits stay with the account.
Is Arumi free to use?
The free browser tools cost nothing. AI generation uses credits. Free credits, such as the sign-up bonus or daily check-in credits, work in the image, video, audio and music generators at each model's lowest settings. Higher settings, paid extras, the canvas and workflows become available with the purchase of a credit pack, a one-time payment. The Pricing page lists the packs and their rules.
Which AI models can I use?
Currently there are 55 models from 8 companies, with video clips up to 30 seconds and output up to 4K. The image, video, audio and music generators each list their models with what each does well, its maximum length and resolution, and the inputs it accepts. The lists follow the live model catalog, so a model appears as soon as it is available to use.
How does billing work?
The Generate button shows how many credits a run costs before it is clicked. If a setting is left to the model, the quote uses the most expensive option the model could choose. A successful generation is charged exactly the quoted credits. Failures caused by us or blocked by our content checks are free. If the model service does not charge for a failure, neither do we. If it does, the charge is for the actual usage, capped at the quote, and the record shows the charge.
What happens if a generation fails?
The card states why the generation failed and whether any credits were charged. Failures caused by us or blocked by our content checks are free. If the model service does not charge for a failure, neither do we. If it does, the charge is for the actual usage, capped at the quote, and the record shows the charge.
Can I use what I make commercially?
As between you and Arumi, the images, video, audio and music generated for you are yours to use, subject to our Terms of Use, the terms of the provider that runs the model, and the law, which in some countries does not grant copyright in machine-generated work. You are responsible for having the rights to anything you upload.
Are my uploads private?
Yes. Photos and files uploaded for a generation or a workflow are private to the account. To produce a result, they are sent to the provider that runs the selected model. The free tools never upload anything at all.
What is the AI canvas?
The AI canvas connects several generation steps into one workflow that can be run as a simple form. Explore lists what people share: creations, with the prompt and settings behind them, which can be made again in one click, and workflows, which can be copied to the canvas to run or change once credits have been purchased.
How do I contact support?
Email support@arumi.ai. Writing from the email address used to sign in allows the account to be found immediately, and replies are sent by email.
Create an image, video, voice-over or song
Choose what to create at the top of this page, write a prompt and check the credits before generating.