pinokio official documentation

Tutorial for Pinokio and its Applications

A specialized fork of SongGeneration Studio, engineered to run on Macs with any M series Apple Silicon chips.

Pinokio script for https://huggingface.co/Ole1/Joy_Caption_Batch-GUI

A lightweight Pinokio wrapper for HeartMuLa/heartlib facilitating one-click local setup and a Gradio UI for music generation.

Pinokio script for @JarodMica's StyleTTS WebUI

Pinokio tool for debugging your peer drives

a pinokio script launcher for Audioghost-AI

FlashPortrait: 6$\times$ Faster Infinite Portrait Animation with Adaptive Latent Prediction For the 10s video (720x1280, fps=25), sequential_cpu_offload requires 10GB VRAM ,model_full_load = 60GB

VoxForge Pro is a premium, offline audiobook generator powered by Kokoro-82M & Chatterbox TTS. Transform PDFs and text into professional audio using 47 lifelike voices across 6 languages. Features include voice cloning, smart OCR for scanned documents, and multi-speaker narration support.

Use PuterOS free credits and models as back-ends to pinokio apps. VERY basic app. More coming soon!!!

MuseTalk is a cutting-edge video-to-video (V2V) lip-sync solution engineered to deliver highly accurate and natural mouth movements synchronized to audio input. Precision LipSync: Realistic and seamless synchronization of speech audio to facial movements. Efficiently designed to run on 8–12 GB VRAM,

High-quality Pinokio desktop UI for the GLM-Image text-to-image and image-to-image model, offering a Gradio interface with advanced controls, auto-saving, and storage optimization for local high-resolution AI image generation.

a pinokio script launcher for Ultimate Vocal Remover 5

Microsoft Edge's online text-to-speech service

IMTalker: Efficient Audio-driven Talking Face Generation with Implicit Motion Transfer. GPU Requirement 12-16gb Vram

UniVideo: Unified Video Understanding, Generation, and Editing — Requires a 24 GB GPU To use this model you need to create a huggingface token by using, Harryji168/univideo-studio when creating token.

SAM 3 is a unified foundation model for promptable segmentation. First, acquire the gated model by visiting https://huggingface.co/facebook/sam3. Once approved, create an HF token and assign it to facebook/sam3. Then, create another token for merve/SAM3-video-segmentation. GPU Requirement 16gb Vram or higher

Expressive Portrait Image Animation for Live Streaming recommended 16–24GB VRAM 16-32gb RAM

Distill and quantize models using TorchAO with intelligent GPU/CPU management. this app was built by Manat, This project is still in its early stages, so feel free to contribute to the build.