Discovered5mo ago
stable-audio-tools
github.com/peanutcocktail

Generative models for conditional audio generation

Discovered5mo ago
langflow
github.com/langflow-ai

Langflow is a powerful tool for building and deploying AI-powered agents and workflows.

Discovered5mo ago
ToonCrafter
github.com/tooncrafter

[SIGGRAPH Asia 2024, Journal Track] ToonCrafter: Generative Cartoon Interpolation

Discovered5mo ago
codebuff
github.com/CodebuffAI

Generate code from the terminal!

Discovered5mo ago
tribev2
github.com/facebookresearch

This repository contains the code to train and evaluate TRIBE v2, a multimodal model for brain response prediction

Discovered5mo ago
SageAttention
github.com/deepbeepmeep

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.

Discovered5mo ago
s2.cpp
github.com/rodrigomatta

Implementation of Fish Audio S2 Pro model inference in native ggml.

Discovered5mo ago
rodrigomt/s2-pro-gguf · Hugging Face
huggingface.co/rodrigomt

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Discovered5mo ago
drbaph/s2-pro-fp8 · Hugging Face
huggingface.co/drbaph

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Discovered5mo ago
slides2video-pinokio
github.com/elloza

Public repository of the slides2video app for pinokio

Discovered5mo ago
Pyramid-Flow
github.com/betapeanut

Code of Pyramidal Flow Matching for Efficient Video Generative Modeling

Discovered5mo ago
FluxRT
github.com/manat0912

Contribute to manat0912/FluxRT development by creating an account on GitHub.

Discovered5mo ago
kepler-website
github.com/Sabriigsxr

Contribute to Sabriigsxr/kepler-website development by creating an account on GitHub.

Discovered5mo ago
scenema-audio
github.com/ScenemaAI

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.

Discovered5mo ago
claude-mem
github.com/thedotmack

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