LogoTopLocal Studio
  • Features
  • FAQ
  • Blog
Download
LogoTopLocal Studio

Your own AI studio, on your own computer.

Ask AI about TopLocal Studio

Product
  • Features
  • Pricing
  • FAQ
Resources
  • Blog
  • GitHub
  • Changelog
Legal
  • Privacy Policy
  • Terms of Service
© 2026 TopLocal Studio. All Rights Reserved.

Changelog

Stay up to date with the latest changes in our product

First release

v0.1.0

The first public version of TopLocal Studio for macOS and Windows

2026/09/24

Four modules

  • Image: text-to-image and editing by instruction, with Z-Image-Turbo, FLUX.2 klein 4B and FLUX.2 klein 9B.
  • Video: text or image to short videos with sound, 3, 5 or 8 seconds at 480p or 720p, with LTX-2.5 Distilled (needs 24 GB+ memory).
  • Music: full songs from one sentence, with your own lyrics or as instrumentals, using ACE-Step 1.5 or YuE2 for Chinese vocals. Includes a built-in lyric writer (Qwen3-4B) and export to a visualizer video.
  • Speech: transcription to TXT and SRT subtitles with Qwen3-ASR 0.6B/1.7B or SenseVoice, and text-to-speech with Kokoro or Qwen3-TTS, including voice cloning.

Working in the app

  • Model picker per task: choose which model to use for each kind of job.
  • Welcome examples: sample prompts to try on first launch.
  • Prompt polish: turn a short idea into a more detailed prompt.
  • Chinese and English UI.

Models and downloads

  • Models are downloaded inside the app when you need them, with resume and sha256 checks.
  • Optional China mirror (hf-mirror.com) for faster downloads in mainland China.
  • Each model shows its license before you download it.

Platforms

  • macOS (Apple Silicon, macOS 13+): Developer ID signed build.
  • Windows 10/11 x64: early build. It is new and has not been widely tested yet, and the installer is not code-signed, so SmartScreen may show a warning.