bfs18

diffusion-posterior-sampling. Official pytorch repository for "Diffusion Posterior Sampling for General Noisy Inverse Problems"

tts. Inworld TTS

viitor-voice-nar. Python

SplatFlow. [CVPR 2025] Official code for the paper "SplatFlow: Multi-View Rectified Flow Model for 3D Gaussian Splatting Synthesis"

OpenMontage. World's first open-source, agentic video production system. 12 pipelines, 52 tools, 500+ agent skills. Turn your AI coding assistant into a full video production studio.

FineEdit. Shell

shanghai-house-data-view. 上海二手房网签数据每日记录 数据来源：网上房地产

shanghai-finance-data-view. 上海住户存贷款数据 数据来源：中国人民银行上海总部官网

minit2i-torch. Official PyTorch re-implementation of MiniT2I.

torchlpc. Fast and differentiable time domain all-pole filter in PyTorch.

dots.tts. Python

LakonLab. Official implementation of AsymFlow, pi-Flow, GMFlow

MOSS-TTS-Nano. MOSS-TTS-Nano is an open-source multilingual tiny speech generation model from MOSI.AI and the OpenMOSS team. With only 0.1B parameters, it is designed for realtime speech generation, can run directly on CPU without a GPU, and keeps the deployment stack simple enough for local demos, web serving, and lightweight product integration.

kokoro. https://hf.co/hexgrad/Kokoro-82M

tinker-cookbook. Post-training with Tinker

nested_learning. A Reproduction of GDM's Nested Learning Paper

FunCineForge. Python

edu-knowlege. 教育各种资料，从幼儿园到小学、中学，涵盖学而思，万维、猿辅导等多个机构，持续增加中

zotmoov. Zotero plugin to automatically move attachments and link them

PLFM_RADAR. Open-source, low-cost 10.5 GHz PLFM phased array RADAR system

andrej-karpathy-skills. A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

SoFlow. [ICLR 2026] SoFlow: Solution Flow Models for One-Step Generative Modeling

traceml. Open-source observability for PyTorch training runs.

telling-left-from-right. Project website for "Telling left from right: Learning spatial correspondence between sight and sound"

sglang-omni. SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models

autoresearch. AI agents running research on single-GPU nanochat training automatically

ironclaw. IronClaw is an Agent OS focused on privacy, security and extensibility

Wechat-YATT. Python

voicebox. The open-source AI voice studio. Clone, dictate, create.

cultivation-world-simulator. 基于 AI Agent 工作流的修仙世界模拟器，旨在还原智能、开放的仙侠世界。| An open-source Cultivation World Simulator using Agentic Workflow to create a dynamic, emerging Xianxia world.

BitDance. BitDance & UniWeTok: Open-source autoregressive model with binary visual tokens. A research project for building powerful multimodal autoregressive model.

FireRedASR2S. A SOTA Industrial-Grade All-in-One ASR system with ASR, VAD, LID, and Punc modules. FireRedASR2 supports Chinese (Mandarin, 20+ dialects/accents), English, code-switching, and both speech and singing ASR. FireRedVAD supports speech/singing/music in 100+ langs. FireRedLID supports 100+ langs and 20+ zh dialects. FireRedPunc supports zh and en.

ijepa. Official codebase for I-JEPA, the Image-based Joint-Embedding Predictive Architecture. First outlined in the CVPR paper, "Self-supervised learning from images with a joint-embedding predictive architecture."

jepa. PyTorch code and models for V-JEPA self-supervised learning from video.

ECC. The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Qwen3-TTS. Qwen3-TTS is an open-source series of TTS models developed by the Qwen team at Alibaba Cloud, supporting stable, expressive, and streaming speech generation, free-form voice design, and vivid voice cloning.

GLM-TTS. GLM-TTS: Controllable & Emotion-Expressive Zero-shot TTS with Multi-Reward Reinforcement Learning

LTX-Video. Official repository for LTX-Video

sam-audio. The repository provides code for running inference with the Meta Segment Anything Audio Model (SAM-Audio), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

armel. poorman's ar-dit tts

cut-cross-entropy. Apple's Cut Cross Entropy

hello-agents. 📚 《从零开始构建智能体》——从零开始的智能体原理与实践教程

audioghost-ai. Extract any sound with text prompts. Memory-optimized SAM-Audio with modern UI.

tcFFT. Cuda

Fun-Audio-Chat. Fun-Audio-Chat is a Large Audio Language Model built for natural, low-latency voice interactions.

FireRedChat. A Fully Self-Hosted Solution for Full-Duplex Voice Interaction

OpenSpec. Spec-driven development (SDD) for AI coding assistants.

lime. AI content workspace for Chinese creators: desktop writing, research, prompt library, knowledge base, and multi-model workflows.

mini-sglang. A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.

banana-slides. 一个基于nano banana pro🍌的原生AI PPT生成应用，迈向＂Vibe PPT＂; 支持上传任意模板图片，上传任意素材&智能解析，一句话/大纲/页面描述自动生成PPT，口头修改指定区域、一键导出可编辑ppt - An AI-native slides generator based on nano banana pro🍌

lws. Fast spectrogram phase recovery using Local Weighted Sums (C/Python/Matlab)

riemannian-fm. code for "Riemannian Flow Matching on General Geometries".

JiT. PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

emotion-annotations. Python

xGeMM. Accelerated General (FP32) Matrix Multiplication from scratch in CUDA

miipher-2. Googleの音声復元モデルMiipher-2の再現実装の学習および推論コード。学習済みモデルも公開しています。

Make-An-Audio-3. Make-An-Audio-3: Transforming Text/Video into Audio via Flow-based Large Diffusion Transformers

MOSS-Speech. MOSS-Speech is a true speech-to-speech large language model without text guidance.

liquid-audio. Liquid Audio - Speech-to-Speech audio models by Liquid AI

kokoro-tts. A CLI text-to-speech tool using the Kokoro model, supporting multiple languages, voices (with blending), and various input formats including EPUB books and PDF documents.

BettaFish. 微舆：人人可用的多Agent舆情分析助手，打破信息茧房，还原舆情原貌，预测未来走向，辅助决策！从0实现，不依赖任何框架。

input-leap. Open-source KVM software

LongCat-Flash-Omni. This is the official repo for the paper "LongCat-Flash-Omni Technical Report"

SoulX-Podcast. SoulX-Podcast is an inference codebase by the Soul AI team for generating high-fidelity podcasts from text.

tread. Python

LightningDiT. [CVPR 2025 Oral] Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models

OpenStock. OpenStock is an open-source alternative to expensive market platforms. Track real-time prices, set personalized alerts, and explore detailed company insights — built openly, for everyone, forever free.

RAE. Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

nanochat. The best ChatGPT that $100 can buy.

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

neutts. On-device TTS model by Neuphonic

cutter. Free and Open Source Reverse Engineering Platform powered by rizin

cpupro. Rethinking of CPU profile analysis

Ming-UniAudio. Ming-UniAudio: Speech LLM for Joint Understanding, Generation and Editing with Unified Representation

tilelang. Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

noFFT. A reference implementation of the Resonate algorithm in C++ for Python.

tinyworlds. A minimal implementation of DeepMind's Genie world model

VoxCPM. VoxCPM2: Tokenizer-Free TTS for Multilingual Speech Generation, Creative Voice Design, and True-to-Life Cloning

Qwen3-Omni. Qwen3-omni is a natively end-to-end, omni-modal LLM developed by the Qwen team at Alibaba Cloud, capable of understanding text, audio, images, and video, as well as generating speech in real time.

tiny-qwen. A minimal PyTorch re-implementation of Qwen 3.5

system-prompts-and-models-of-ai-tools. FULL Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI, VSCode Agent, Warp.dev, Windsurf, Xcode, Z.ai Code, Dia & v0. (And other Open Sourced) System Prompts, Internal Tools & AI Models

ACE. NVIDIA ACE samples, workflows, and resources

index-tts. An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

term.everything. Run any GUI app in the terminal❗

WhisperLiveKit. Simultaneous speech-to-text models

panda_quantflow. JavaScript

winboat. Run Windows apps on 🐧 Linux with ✨ seamless integration

WenetSpeech-Yue. A Large-scale Cantonese Speech Corpus with Multi-dimensional Annotation

ComfyUI. The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

Step-Audio2. Step-Audio 2 is an end-to-end multi-modal large language model designed for industry-strength audio understanding and speech conversation.

script-doctor. Code for PuzzleJAX, a benchmark for reasoning and learning, that reimplements PuzzleScript, a concise and expressive DSL and game engine for grid-based puzzle games, in JAX.

bournemouth-forced-aligner. Extract phoneme-level timestamps from speeh audio.

teach-fish-to-swim. Wanna breeze through some papers?

f2. High-speed downloader for multiple platforms

torchcodec. PyTorch media decoding and encoding

whisper-finetune. Fine-tune and evaluate Whisper models for Automatic Speech Recognition (ASR) on custom datasets or datasets from huggingface.

tiny-tpu. A minimal tensor processing unit (TPU), inspired by Google's TPU V2 and V1

VibeVoice. Open-Source Frontier Voice AI

discrete-fm. Educational implementation of the Discrete Flow Matching paper

MaAI. A real-time software for turn-taking, backchannel, and head-nodding prediction

RLFromScratch. Python

FlashCosyVoice. FlashCosyVoice: A lightweight vLLM implementation built from scratch for CosyVoice.

DiariZen. A toolkit for speaker diarization.

whisper-diarization. Automatic Speech Recognition with Speaker Diarization based on OpenAI Whisper

train-higgs-audio. Text-audio foundation model from Boson AI

Marco-Voice. A Unified Framework for Expressive Speech Synthesis with Voice Cloning

MeanAudio. [ACL 2026 Main] MeanAudio: Fast and Faithful Text-to-Audio Generation with Mean Flows

Emilia-NV. Official Repository of Paper: "Emilia-NV: A Non-Verbal Speech Dataset with Word-Level Annotation for Human-Like Speech Modeling"

KittenTTS. State-of-the-art TTS model under 25MB 😻

gpt-oss. gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

Nes2Net_ASVspoof_ITW. Python

Nes2Net. Python

ClearerVoice-Studio. An AI-Powered Speech Processing Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Enhancement, Separation, and Target Speaker Extraction, etc.

HRM. Hierarchical Reasoning Model Official Release

X-Omni. Official inference code and LongText-Bench benchmark for our paper X-Omni (https://arxiv.org/pdf/2507.22058).

SpeechSplit. Unsupervised Speech Decomposition Via Triple Information Bottleneck

EmergentTTS-Eval-public. [NeurIPS' 25] Benchmark for evaluating TTS models on complex prosodic, expressiveness, and linguistic challenges.

higgs-audio. Text-audio foundation model from Boson AI

VoiceBench. [TACL'26] VoiceBench: Benchmarking LLM-Based Voice Assistants

C-Optim. [ICLR 2026] When it comes to optimizers, it's always better to be safe than sorry

DreamOn. Diffusion Language Models For Code Infilling Beyond Fixed-size Canvas

ai-engineering-hub. In-depth tutorials on LLMs, RAGs and real-world AI agent applications.

freeCodeCamp. freeCodeCamp.org's open-source codebase and curriculum. Learn math, programming, and computer science for free.

audio-flamingo. PyTorch implementation of Audio Flamingo: Series of Advanced Audio Understanding Language Models

Neuro. A recreation of Neuro-Sama originally created in 7 days.

ZipVoice. Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching

audioset_tagging_cnn. Python

Bella. Bella is best

doccano. Open source annotation tool for machine learning practitioners.

zhvoice. Chinese voice corpus. 中文语音语料，语音更加清晰自然，包含8个开源数据集，3200个说话人，900小时语音，1300万字。

Long-RL. Long-RL: Scaling RL to Long Sequences (NeurIPS 2025)

torch-profiling-tutorial. Python

speech-to-phrase. Fast and personalized local speech-to-text

MuCodec. Python

BiliTools. 本项目已停止维护。

community-events. Place where folks can contribute to 🤗 community events

nano-vllm. Nano vLLM

CrossFlow. [CVPR2025] PyTorch-based reimplementation of CrossFlow, as proposed in 'Flowing from Words to Pixels: A Noise-Free Framework for Cross-Modality Evolution'

TTSizer. 🎙️ Automatically transcribe audio/video into high-quality, speaker-specific Text-To-Speech datasets

vosk-api. Offline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node

adjoint_sampling. code for "Adjoint Sampling: Highly Scalable Diffusion Samplers via Adjoint Matching"

chatterbox. SoTA open-source TTS

Dolphin. Dolphin is a multilingual, multitask ASR model jointly trained by DataoceanAI and Tsinghua University.

Megakernels. Kernels, of the mega variety :)

RelativisticGAN. Code for replication of the paper "The relativistic discriminator: a key element missing from standard GAN"

BigCodec. Official implementation of the paper "BigCodec: Pushing the Limits of Low-Bitrate Neural Speech Codec"

MeanFlow. PyTorch implementation of MeanFlow & iMF (one-step generative modeling).

SLED-TTS. Streamable Text-to-Speech model using a language modeling approach, without vector quantization

ten-vad. Voice Activity Detector (VAD) : low-latency, high-performance and lightweight

AdjointDPM. Python

browserbee. 🐝 AI-powered browser assistant ("Cline for web browsing")

AgentCPM-GUI. AgentCPM-GUI: An on-device GUI agent for operating Android apps, enhancing reasoning ability with reinforcement fine-tuning for efficient task execution.

LACT. Linux GPU Configuration And Monitoring Tool

tracy. Frame profiler

flow_grpo. [NeurIPS 2025] An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL

VITA-Audio. ✨✨[NeurIPS 2025] VITA-Audio: Fast Interleaved Cross-Modal Token Generation for Efficient Large Speech-Language Model

nanoVLM. The simplest, fastest repository for training/finetuning small-sized VLMs.

RealtimeVoiceChat. Have a natural, spoken conversation with AI!

DANSM. [ICLR 2025 spotlight]: Easing Training Process of Rectified Flow Models Via Lengthening Inter-Path Distance

MiMo. MiMo: Unlocking the Reasoning Potential of Language Model – From Pretraining to Posttraining

audioflow. Python

Muyan-TTS. Python

Kimi-Audio. Kimi-Audio, an open-source audio foundation model excelling in audio understanding, generation, and conversation

dia. A TTS model capable of generating ultra-realistic dialogue in one pass.

TouchNet. A native-PyTorch library for large scale M-LLM (text/audio) training with tp/cp/dp.

Tiny-GRPO. minimal GRPO implementation from scratch

minimind. 🧠「大模型」2小时完全从0训练64M的小参数LLM！Train a 64M-parameter LLM from scratch in just 2h!

mobilerun. Automate your mobile devices with natural language commands - an LLM agnostic mobile Agent 🤖

openai-fm. Code for openai.fm, a demo for the OpenAI Speech API

ableton-mcp. Python

ttt-video-dit. Official PyTorch implementation of One-Minute Video Generation with Test-Time Training

async_cosyvoice. 使用vllm加速cosyvoice2的推理

sglang. SGLang is a high-performance serving framework for large language models and multimodal models.

nano-aha-moment. Single File, Single GPU, From Scratch, Efficient, Full Parameter Tuning library for "RL for LLMs"

RealtimeTTS. Converts text to speech in realtime

moshi-finetune. Python

LeetCUDA. 📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.🎉

airi. 💖🧸 Self hosted, you-owned Grok Companion, a container of souls of waifu, cyber livings to bring them into our worlds, wishing to achieve Neuro-sama's altitude. Capable of realtime voice chat, Minecraft, Factorio playing. Web / macOS / Windows supported.

MCP-Chinese-Getting-Started-Guide. Model Context Protocol(MCP) 编程极速入门

Orpheus-FastAPI. High-performance Text-to-Speech server with OpenAI-compatible API, 8 voices, emotion tags, and modern web UI. Optimized for RTX GPUs.

Qwen2.5-Omni. Qwen2.5-Omni is an end-to-end multimodal model by Qwen team at Alibaba Cloud, capable of understanding text, audio, vision, video, and performing real-time speech generation.

csm. A Conversational Speech Generation Model

shortcut-models. Python

vocoder. Python

understand-r1-zero. Understanding R1-Zero-Like Training: A Critical Perspective

mine-pytorch. Mutual Information Neural Estimation in Pytorch

VoxInstruct. VoxInstruct: Expressive Human Instruction-to-Speech Generation with Unified Multilingual Codec Language Modelling

LLMs-from-scratch. Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

imm. Official implementation of Inductive Moment Matching

r1-aqa. 🤗 R1-AQA Model: mispeech/r1-aqa

ai_quant_trade. 股票AI操盘手：从学习、模拟到实盘，一站式平台。包含股票知识、策略实例、大模型、因子挖掘、传统策略、机器学习、深度学习、强化学习、图网络、高频交易、C++部署和聚宽实例代码等，可以方便学习、模拟及实盘交易

r1_vlm. Build your own visual reasoning model

fire_red_agent. Ruby

OpenManus. No fortress, purely open ground. OpenManus is Coming.

lingua-py. The most accurate natural language detection library for Python, suitable for short text and mixed-language text

Awesome-LLM-Post-training. Awesome Reasoning LLM Tutorial/Survey/Guide

Grokking-the-Coding-Interview-Patterns-for-Coding-Questions. Grokking the Coding Interview: Patterns for Coding Questions Alternative

bit-brain. 最少使用 3090 即可训练自己的比特大脑（miniLLM）🧠（进行中）. Train your own BitBrain(A mini LLM) with just an RTX 3090 minimum.

LLMs-Zero-to-Hero. 从无名小卒到大模型（LLM）大英雄~ 欢迎关注后续！！！

PPO-PyTorch. Minimal implementation of clipped objective Proximal Policy Optimization (PPO) in PyTorch