Vincent Gourbin

I get the latest AI models running on Apple silicon, and turn them into apps that need no server.

I port models released by AI labs, image, video, speech, music, to Swift and MLX. I publish those ports as open source, then build them into apps for Mac, iPhone and iPad. Your data stays on your device.

The forge

Each row follows one model, from its release to the app that uses it.

  1. Released model : FLUX.2Black Forest Labs
    Swift / MLX port : flux-2-swift-mlx43 stars
    In the app : Fluxforge Studioimages
  2. Released model : LTX-2Lightricks
    Swift / MLX port : ltx-video-swift-mlx28 stars
    In the app : Fluxforge Studiovideo
  3. Released model : Gemma 4Google
    Swift / MLX port : gemma-4-swift-mlx37 stars
    In the app : Fluxforge Studiovideo encoder and prompts
  4. Released model : YuE2Multimodal Art Projection
    Swift / MLX port : yue2-mlx-swift1 star
    In the app : PocketAnthemsongs
  5. Released model : VoxtralMistral AI
    Swift / MLX port : mlx-voxtral-swift6 stars
    In the app : Meeting CompAIniontranscription
  6. Released model : RMBG 2.0BRIA AI
    Swift / MLX port : RMBG2Swift5 stars
    In the app : Fluxforge Studiobackground removal
  7. Released model : Qwen3.8Alibaba Qwen
    Swift / MLX port : qwen38-mlx-swift
    In the app : not in an app yet
  8. Released model : MiniMax-H3MiniMax
    Swift / MLX port : minimax-h3-swift-mlx1 star
    In the app : not in an app yet

Apps

Four apps, all computed on the device.

Log

Collected automatically from GitHub, Hugging Face and the App Store on 12 October 2026.

  1. mlx-voxtral-swift

    Native Swift implementation of Voxtral speech-to-text for Apple Silicon with MLX

  2. yue2-mlx-swift

    Swift/MLX port of YuE2 (m-a-p/YuE2-3B + YuE2-Vae) for Apple Silicon: lyrics + style -> song, on Mac and iPhone. Powers PocketAnthem on the App Store.

  3. VincentGOURBIN/yue2-mlx-packsnew

    Model published on Hugging Face

  4. VincentGOURBIN/sheetsage2-mlx-q4new

    Model published on Hugging Face

  5. gemma-4-swift-mlx

    Native Gemma 4 multimodal (text+vision+audio+video) inference for Apple Silicon via MLX Swift

  6. qwen38-mlx-swift

    Native Qwen3.8 inference on Apple Silicon with MLX Swift — dense 27B, Bonsai 2 and Flash-Next MoE; embeddable Qwen38Brain library (fast/lean profiles, conversation reuse), OpenAI-compatible server with tool calling, LangWatch-benchmarked agent loop

  7. swift-mlx-profiler

    MLX performance profiling for Apple Silicon — GPU/CPU utilization, Chrome Trace (Perfetto), benchmarks

  8. Fluxforge Studio 3.2

    New version on the App Store

  9. flux-2-swift-mlx

    Flux.2 image generation on Mac with MLX - Native Swift implementation

Open source

The most followed ports on GitHub.

  • flux-2-swift-mlx

    Flux.2 image generation on Mac with MLX - Native Swift implementation

    Swift, 43 stars

  • gemma-4-swift-mlx

    Native Gemma 4 multimodal (text+vision+audio+video) inference for Apple Silicon via MLX Swift

    Swift, 37 stars

  • ltx-video-swift-mlx

    LTX-2 text-to-video generation for Apple Silicon using MLX Swift

    Swift, 28 stars

  • yue2-mlx-swift

    Swift/MLX port of YuE2 (m-a-p/YuE2-3B + YuE2-Vae) for Apple Silicon: lyrics + style -> song, on Mac and iPhone. Powers PocketAnthem on the App Store.

    Swift, 1 star

  • swift-mlx-profiler

    MLX performance profiling for Apple Silicon — GPU/CPU utilization, Chrome Trace (Perfetto), benchmarks

    Swift

  • mlx-voxtral-swift

    Native Swift implementation of Voxtral speech-to-text for Apple Silicon with MLX

    Swift, 6 stars

The full catalogue: 22 repositories, 27 models, datasets and demos

Support

The ports are free to use and three of the four apps are free. If one of them helped you, a coffee helps me keep going.

Buy me a coffee on Buy Me a Coffee