Multi-Voice Text-to-Speech for Stories and Audiobooks. Supports Kokoro and Chatterbox TTS engines with GPU acceleration.
Radar
New projects people are discovering or following across Pinokio.
Branded text-to-video studio for long-form documentary YouTube videos. Local LLM scripts via LM Studio, web image sourcing with lightweight AI recreation (uniform 1080p, watermark/text removal), local Kokoro TTS narration, cinematic Ken Burns rendering, automatic GPU load/unload, and a modern web UI to manage everything.
Super fast Multilingual TTS supporting 54 voices across 8 languages.
[NVIDIA ONLY] YuEGP--A Web UI for YuE, an Open Full-song Generation Foundation Model (10G VRAM required), via https://github.com/deepbeepmeep/YuEGP
A multi-voice AI audiobook generator built on Qwen3-TTS — annotate scripts with an LLM, assign unique voices to each character, per-line style instructions for delivery, clone voices from reference audio, design new voices from text descriptions, train custom voices with LoRA fine-tuning, and export to MP3 or Audacity multi-track projects
Text-to-Video (T2V) generation framework from Vchitect https://github.com/Vchitect/LaVie
AutoGPT is a powerful tool that lets you create and run intelligent agents https://github.com/Significant-Gravitas/AutoGPT
Open-source Facebook Marketplace scraper in Python. Export listing title, price, location, listed date and photo to JSON or CSV.
Artist is a training-free text-driven image stylization method. You give an image and input a prompt describing the desired style, Artist give you the stylized image in that style. The detail of the original image and the style you provide is harmonically integrated https://huggingface.co/spaces/fffiloni/Artist

Private, local image and PDF OCR with your choice of LightOnOCR-3 0.8B, 1B or 4B. NVIDIA, Apple Silicon and CPU support.
State-of-the-art 1B OCR model (83.2% on OlmOCR-Bench). Local version of the HuggingFace demo. Created by Claude Code, orchestrated by TheAwakenOne.
[v0.5.1] FramePack Video App offering multiple generation types: Original, F1, video extension, end frame. Features include: LoRA support, job queueing, advanced timestamped prompts, offline mode, a post-processing suite including upscaling, interpolation, filters and more!