Run Kokoro TTS
Ready in about 90 seconds. No install, no setup. From $0.22 per hour.
What it does
Kokoro turns written text into spoken audio. It ships with a set of voices in several accents and it sounds close to a real narrator.
It is remarkably small for what it does, which makes it fast and cheap. An hour of finished narration costs a few cents of GPU time.
It is the right tool for video narration, audio versions of articles, product demos, and accessibility work. It is not a voice cloning tool and will not copy a specific person.
What you can make
These are prompts, not fixed presets. The workspace opens with one of them already loaded so your first result is one click away.
Your setup
We run this on an RTX 3090. It generates roughly 60 seconds of speech per second of GPU time, so an hour of narration takes about a minute. $0.22 per hour, billed by the minute.
Why this card?
At 82 million parameters this is one of the smallest useful models on the site. It runs on almost anything. We put it on the cheapest card in the fleet because anything faster would finish the job before it finished starting up, and you would pay more for the privilege.
- Memory
- 24GB
- Speed
- about 1 second per minute of narration
- Rate
- $0.22/hr
What will it cost you?
Billed by the minute on a RTX 3090 at $0.22 per hour. A batch job shuts the GPU off when the last minute of narration is done, so this is the whole cost.
Two ways to run it
Click and use it
Paste text, pick a voice, listen, adjust, download the audio.
Get early access Best for exploring.Hand us the whole pile
Send a spreadsheet of scripts and collect one audio file per row.
Join the list Best for volume. The GPU shuts off automatically when it is done.Model facts
| Parameters | 82M |
|---|---|
| License | Apache 2.0 |
| Memory required | 2GB minimum |
| Base model | StyleTTS 2 architecture |
| Publisher | hexgrad |
| Recommended hardware | RTX 3090, 24GB |
| Hugging Face | hexgrad/Kokoro-82M |
Want Kokoro TTS the day it opens?
We are testing the rental flow with a small group first. Leave an email and we will tell you when Kokoro TTS is ready to run. One message, no newsletter.
Kokoro TTS questions
Do I need to install anything to use Kokoro TTS?
No. We start a machine with Kokoro TTS already loaded and hand you a link. Everything runs in your browser, there is nothing to download, and nothing is left on your computer afterwards. It works the same on a Mac, a Windows laptop, or a Chromebook. A workspace is usually ready in about 90 seconds.
How much does it cost to run 120 minutes of narration?
About under $0.01. Kokoro TTS takes about 1 second per minute of narration on the RTX 3090 we recommend, so 120 minutes of narration is roughly 2 minutes of GPU time at $0.22 per hour. Billing is by the minute, and a batch job shuts the GPU off the moment the last item finishes. The calculator above works this out for your own numbers.
Can I run Kokoro TTS on a cheaper card?
Sometimes, and the calculator will not always make it look worth it. At 82 million parameters this is one of the smallest useful models on the site. It runs on almost anything. We put it on the cheapest card in the fleet because anything faster would finish the job before it finished starting up, and you would pay more for the privilege. If you want to try a different card anyway, the advanced catalog lets you pick one and shows the estimated time before you commit.
Can I use Kokoro TTS commercially?
Yes. Kokoro TTS is released under the Apache 2.0 license, which permits commercial use with no conditions worth worrying about. You can use the output in paid client work and in products you sell.
What happens to my files after I stop?
Anything you save into your workspace folder stays in your account and is there when you come back. Everything else is destroyed: the machine is terminated and its disk is wiped. Batch job inputs and outputs are deleted from our storage 24 hours after you download them. We do not read your files or use them to train anything.
What if it fails partway through?
You are not charged for work that did not complete, and failed sessions are refunded automatically. Batch jobs write each output as it finishes rather than at the end, so a failure at item 380 of 400 leaves you 379 usable files and a message saying what broke. Restarting picks up where it stopped rather than redoing the work.
Own a GPU that sits idle? Put it to work. The average provider earns $180 to $420 per month per card.
List your GPU