Kokoro-FastAPI-GPU

Kokoro-FastAPI is a Dockerized FastAPI server that wraps the Kokoro-82M text-to-speech model for local, OpenAI-compatible speech generation.

  • AI
  • Other
  • Productivity
  • Tools & Utilities
Kokoro-FastAPI is a Dockerized FastAPI server that wraps the Kokoro-82M text-to-speech model for local, OpenAI-compatible speech generation. This NVIDIA GPU build uses PyTorch to synthesize multi-language audio with automatic segment stitching, aimed at homelabbers who want high-quality on-device speech.

Platform availability

Available on 1 of 10 platforms

  • Unraid
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)
  • Any server (not listed)

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

62/100 Good

Maintained
100last commit 8 days ago
Popular
755,682 GitHub stars
Easy to install
351 install platform, install notes, Docker image
Light to run
304 GB minimum RAM
Security check
201 serious finding, 4 minor, CI, tests

Availability: platform listings confirmed today, links checked 21 days ago

GitHub stars
5,682
Open issues
10
Last commit
2026-11-13
Latest release
v0.10.0
License
Apache-2.0
Activity
rising

Checked 7 days ago - source: GitHub (remsky/Kokoro-FastAPI)

Resources & Compatibility

Minimum requirements: 4 GB RAM, 2 CPU core(s).

Checked 54 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Install the Kokoro library first Run `npm i kokoro-js` in the project before starting the app. This installs the speech code the app needs, and without it the program can stop with a module-not-found error. source
  • Pick the right device and precision When loading the model, choose the compute device (where the math happens) that matches your hardware, such as webgpu or cpu, and a data type like fp32 or q8. A mismatch can make startup fail or speech run much slower than expected. source
  • Set a valid model ID and voice Use a valid model name (the Hugging Face model ID in the docs) and an available voice name, such as af_heart, when generating audio. If either value is wrong, the app will not be able to create speech. source

Fetched 53 days ago - each note links to its source

Alternatives

No alternatives collected yet.

Common questions

What port does the Dockerized Kokoro-FastAPI service expose?

The service runs on port 8880.

What Linux GPU issue should non-root deployments watch for?

Some Linux users may hit GPU permission issues when running as non-root. The troubleshooting docs suggest common fixes and note security tradeoffs.

Which HTTP endpoint does the wrapper expose for text-to-speech?

It exposes an OpenAI-compatible endpoint at /v1/audio/speech.

Is there an Unraid-specific support thread for this Dockerized wrapper?

Yes, Unraid has a support thread for FastKoko, the Dockerized Kokoro-FastAPI wrapper.

Answers sourced from deepwiki.com, forums.unraid.net, github.com, localaimaster.com

Community

Collected 49 days ago - public community threads