Quantized Attention that achieves speedups of 2.1-3.1x and 2.7-5.1x compared to FlashAttention2 and xformers, respectively, without lossing end-to-end metrics across various models.
Radar
New projects people are discovering or following across Pinokio.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Code of Pyramidal Flow Matching for Efficient Video Generative Modeling
Contribute to Sabriigsxr/kepler-website development by creating an account on GitHub.
Zero-shot expressive voice cloning and speech generation. Generate anything from short clips to full-length audiobooks with realistic emotional delivery, pacing, and breath control. Clone any voice from a 10-second reference and perform emotions the original speaker never recorded.
Zero-Shot Material Transfer from a Single Image
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Contribute to b00y0h/claude-devbox development by creating an account on GitHub.
Contribute to manat0912/Face-Animation--facevid-2vid development by creating an account on GitHub.
Open-source Claude Design alternative. One-click import your Claude Code / Codex API key. Prompt → prototype / slides / PDF. Multi-model (Claude, GPT, Gemini, Kimi, GLM, Ollama). BYOK, local-first, MIT.
Enter a description of the picture you want, choose an aspect ratio and optional settings, and the app creates the image for you. The generated picture is shown on the screen and can be downloaded....
