Lekh AI
Pro
The complete on-device AI studio for your Mac. Generate images, videos, and music; run local LLMs; transcribe audio; benchmark models; and convert Hugging Face models to MLX, GGUF, or JANG. No cloud, no API keys.
Everything in One App
Lekh AI Pro includes every local AI tool unrestricted: private AI chat, offline content creation, model conversion, benchmarking, STT, TTS, RAG, and creative generation.
Flux Image Generation
State-of-the-art Flux models for stunning AI image creation directly on your Mac
LTX 2.3 Video Generation
Generate AI videos from text or images with LTX 2.3. Up to 1024×768 with synchronized audio, all on-device
On-Device AI Chat
Private conversations with 1000+ language models running entirely on your Mac
AI Image Editing
Inpainting with Flux and intelligent editing with Qwen. Erase, modify, and transform images on-device
Text-to-Speech
Natural voices with Kokoro and Qwen3 TTS. Create audiobooks from any document
Audiobooks
Import, read, and interact with PDFs and EPUBs using AI-powered comprehension
Image Upscaling
4x AI upscaling with RealESRGAN. Enhance any image to stunning resolution
Knowledge Hub
RAG-powered document retrieval. Build personal knowledge bases with vector search
WhisperKit STT
Accurate speech-to-text powered by WhisperKit. Transcribe audio locally with Apple Silicon acceleration
ACE-Step Music Generation
Generate full songs with vocals and instruments using ACE-Step 1.5 XL. Text-to-music, entirely on-device
Safetensor Pipeline
Load custom SD and SDXL safetensor models with LoRA support. Full control over your image generation pipeline
Model Converter
Convert Hugging Face models to MLX, GGUF, or JANG format directly in-app. Custom quantization with one click
Benchmarking
Measure tokens per second, time to first token, and memory usage. Compare models side by side
Created with Lekh AI Pro
All images generated entirely on-device using Flux models. Videos powered by LTX 2.3. No cloud, no limits.
Image to Video
Turn any image into a video with LTX 2.3. All processing happens locally on your Mac.
AI Music Generation
Generate full songs with vocals and instruments using ACE-Step 1.5 XL. All on your Mac, no cloud required.
Demo Track 1
Generated with ACE-Step 1.5 XL
Demo Track 2
Generated with ACE-Step 1.5 XL
Convert & Benchmark
Convert any Hugging Face model and benchmark local AI performance - all built into the app.
Model Converter
Convert any Hugging Face model to MLX, GGUF, or JANG format directly in-app. Choose your quantization level, hit convert, and start chatting - no terminal needed.
Benchmarking
Measure tokens per second, time to first token, and peak memory usage. Compare models head-to-head to find the best fit for your hardware.
Benchmarking in Action
Why Pro?
🎨 Flux Models
Generate stunning images with Flux Dev, Flux Schnell, and Flux2 Klein models. Not available in the App Store version due to platform restrictions.
🎬 LTX 2.3 Video
Generate AI videos from text or images with LTX 2.3. Up to 1024×768 with synchronized audio — all running locally on your Apple Silicon Mac.
🎵 ACE-Step Music
Generate full songs with vocals and instruments using ACE-Step 1.5 XL. Text-to-music in 50+ languages, running entirely on your Mac.
🔓 No Sandbox
Direct distribution means no App Store sandbox restrictions. Full system access for running mflux and other advanced AI tools natively.
⚡ Same Great App
AI chat, image generation, video generation, music generation, TTS, knowledge hub, translation, upscaling — plus Pro-exclusive features. Mac only.
🔒 Still Private
100% on-device processing. No data ever leaves your Mac. Your conversations, images, videos, music, and documents stay completely private.
Get Lekh AI Pro
One-time purchase. No subscription. Yours forever.
Purchase your license below via secure Stripe checkout.
After payment, your license file downloads instantly.
Download the app, open it, and enter your email and license key. That's it.
Secure checkout powered by Stripe · Requires macOS 15+ & Apple Silicon
Note: Lekh AI Pro is a Mac-only purchase. The iOS app is available separately on the App Store.
Already purchased? Open the app and enter your email and license key.