AI Awesome

• LLM Awesome
• ML Awesome
• Stable-Diffusion Awesome
• Voice
  • jianchang512/clone-voice
  • FunAudioLLM/CosyVoice
    • ASR, TTS
  • wenet-e2e/wenet
    • STT
• API/Aggregator/Gateway
  • langfuse/langfuse
    • MIT+EE, TS, React
    • LLM Observability, Prompt Management, LLM Evaluations, Datasets, LLM Metrics, Prompt Playground
  • https://openrouter.io
  • https://together.ai
• Search/RAG
  • Perplexity
  • ItzCrazyKns/Perplexica
    • MIT, TS
    • AI-powered search engine
    • alternative to Perplexity
  • zaidmukaddam/scira
    • MIT, TS
    • MiniPerplx -> scira
    • minimalistic AI-powered search engine
• Client/Desktop/App
  • chatbox
    • GPLv3, TypeScript, React, Electron
    • 支持 Desktop, iOS, Android, Web
  • mckaywrigley/chatbot-ui
    • MIT
  • CherryHQ/cherry-studio
    • Apache-2.0, TS
  • mainframecomputer/fullmoon-ios
    • MIT, Swift
  • languine-ai/languine
    • AGPLv3, TS
    • CLI, 用于 i10n
• WebUI/Chatbot
  • vercel/ai-chatbot
    • MIT, React
  • Yidadaa/ChatGPT-Next-Web
    • MIT, TS, React
  • nluxai/nlux
    • MPL-2.0, TS, React
    • UI for any LLM, supporting LangChain / HuggingFace / Vercel AI,
  • CopilotKit
    • MIT, TS
    • React UI + elegant infrastructure for AI Copilots
    • framework for building custom AI Copilots
    • npm:@copilotkit/react-core,@copilotkit/react-ui,@copilotkit/react-textarea
  • Oneirocom/Magick
    • toolkit for AI builder
  • hubtype/botonic
    • MIT, TS, React
  • open-webui/open-webui
    • BSD-3, Svete, Python
    • WebUI for LLMs
    • open-webui/desktop
      • BSD-3, TS, Svelte, Electron
  • lobehub/lobe-chat
    • Apache-2.0, TypeScript, React
  • yoctol/bottender
    • MIT, TS, React
• refly-ai/refly
  • Apache-2.0, TS
• Document
  • deepdoctection/deepdoctection
  • table-transformer-structure-recognition
  • Table Extraction using LLMs: Unlocking Structured Data from Documents
  • opendatalab/DocLayout-YOLO
  • microsoft/table-transformer
    • TATR
  • phamquiluan/jdeskew
    • MIT, Python
    • ICIP 2022: Adaptive Radial Projection on Fourier Magnitude Spectrum for Document Image Skew Estimation
  • clovaai/donut
    • MIT, Python
    • OCR-free Document Understanding Transformer (Donut) and Synthetic Document Generator (SynthDoG), ECCV 2022
• Text
  • facebookresearch/fastText
    • MIT, C++
    • Word representation and sentence classification
• Image
  • franciszzj/Leffa
    • MIT, Python
    • CVPR 2025 Learning Flow Fields in Attention for Controllable Person Image Generation
• Audio
  • sergree/matchering
    • GPLv3, Python
    • Audio Matching and Mastering
• OCR
  • getomni-ai/zerox
    • MIT, TS, Python
    • OCR & Document Extraction using vision models
    • https://getomni.ai/ocr-benchmark
  • allenai/olmocr
    • Toolkit for linearizing PDFs for LLM datasets/training
• Provider
  • Google Colab
  • Runpod
  • Autodl
• Editor
  • cursor
  • windsurf
    • by Codex
    • OpenAI 收购
  • Aider
  • cline/cline
  • RooVetGit/Roo-Code
    • fork cline
• Eval
  • stanford-crfm/helm
• microsoft/unilm
  • MIT, Python
  • https://thegenerality.com/agi/
• Mintplex-Labs/anything-llm
  • MIT, JS
  • all-in-one Desktop & Docker AI application with built-in RAG, AI agents, No-code agent builder,
• https://github.com/NielsRogge/Transformers-Tutorials

应用

产品/服务

• Chat
  • https://chat.openai.com
  • https://bard.google.com
  • https://claude.ai
• https://bing.com/create
• https://beta.dreamstudio.ai/generate
• https://playgroundai.com/
• chatpdf
  • 阅读 PDF
• roomgpt
  • 生成房间，家装
• Chat GPT 应用
  • https://open-gpt.app/
  • 万能道歉信
• design/image/ppt
  • dalle
    • https://dalle.party/
      • https://news.ycombinator.com/item?id=38432486
  • https://designer.microsoft.com/
  • https://pixian.ai/
    • Remove Image Backgrounds
  • https://quickqr.art/
    • QR Code Art
  • https://ideogram.ai/
  • https://github.com/diff-usion/Awesome-Diffusion-Models
• SVG/vector
  • https://logodiffusion.com/
  • https://github.com/GeorgLegato/stable-diffusion-webui-vectorstudio
  • https://vectorart.ai/
  • https://vectormagic.com/

---

• openrecall/openrecall
  • AGPLv3, Python
  • Windows 11 Recall 替代
• williamfzc/chat-gpt-ppt
  • 生成 PPT
• mckaywrigley/ai-code-translator
  • 编程语言转换
• whitead/paper-qa
  • 文档理解，QA
• mpaepper/content-chatbot
  • 网站内容, QA
• yetone/openai-translator
  • 翻译
• microsoft/visual-chatgpt
  • 图片编辑
• hwchase17/langchain
• LAION-AI/Open-Assistant
  • 助手
• togethercomputer/OpenChatKit
• gpt-3-experiments
• bhaskatripathi/pdfGPT
• freedmand/semantra
  • semantic search
• https://anonymous-31415926.github.io/
• 发现
  • https://www.theaivalley.com/
• ShishirPatil/gorilla
• alesaccoia/VoiceStreamAI
  • Near-Realtime audio transcription using self-hosted Whisper and WebSocket in Python/JS
• SJTU-IPADS/PowerInfer
  • High-speed Large Language Model Serving on PCs with Consumer-grade GPUs
• Portkey-AI/gateway
  • MIT, Typescript
  • Blazing Fast AI Gateway. Route to 100+ LLMs with 1 fast & friendly API
• 字幕
  • melon/qingwu-zimu
    • whisper

Coding

tip

• 功能方向
  • Chat - 基于代码询问、解释、找 Bug
  • 编辑 - 基于 Prompt 修改选中代码
  • Completion - 智能代码补全
  • Generation - 生成代码
  • 功能场景生成 - Commit Message、Commit Explain、Issue、PR、文档、注释
  • NLP -> Code/Query/BI
  • NLP/Vision -> Design/Image/Code(HTML+CSS)
  • 编辑器集成

• continuedev/continue
  • Apache-2.0, Python
  • ChatGPT -> VS Code and JetBrains
• THUDM/CodeGeeX
  • Apache-2.0
  • Multilingual Code Generation Model
• cloudquery/pg_gpt
  • 自然语言查询 Pg
• codota/TabNine
  • 支持 VSC 和 JB
  • codota/tabnine-intellij
    • Self host 收费
• CodedotAl/gpt-code-clippy
  • 类似 GitHub Copilot
  • HN
• 不开源/商业版
  • https://github.com/gitbito/bitoai
    • Jetbrains 插件

Reading

• https://github.com/microsoft/generative-ai-for-beginners
• https://www.promptingguide.ai/zh/readings
• https://jalammar.github.io/
• https://karpathy.ai/zero-to-hero.html
• https://github.com/premAI-io/state-of-open-source-ai

Prompt

• https://huggingface.co/spaces/merve/ChatGPT-prompt-generator
• prompt-engineering/click-prompt
• https://sharegpt.com/

Vision

• 换脸
  • Deepfake
• 骨骼
• 编辑
• 对嘴形
• 商业
  • Heygen
  • http://Rask.AI
• https://news.ycombinator.com/item?id=35352452
  • Text2Video
  • https://github.com/Picsart-AI-Research/Text2Video-Zero
• tldraw/make-real
  • Draw -> HTML
  • https://makereal.tldraw.com/
    • https://github.com/tldraw/make-real
  • https://www.tldraw.com/
• excalidraw/excalidraw
  • text -> diagram
• OthersideAI/self-operating-computer
  • text -> 电脑自动操作
• https://github.com/Azure/gen-cv/tree/main/avatar/interactive
  • https://techcommunity.microsoft.com/t5/ai-azure-ai-services-blog/azure-ai-speech-announces-public-preview-of-text-to-speech/ba-p/3981448

UI

Playground

• oobabooga/text-generation-webui
  • AGPLv3, Python
  • LLaMA, llama.cpp, GPT-J, OPT, GALACTICA
• AUTOMATIC1111/stable-diffusion-webui
  • AGPLv3, Python
• nat/openplayground
  • https://nat.dev/
• webonnx/wonnx
  • MIT
  • ONNX
• AbdBarho/stable-diffusion-webui-docker
• LostRuins/koboldcpp

---

• lxe/simple-llama-finetuner
  • LLaMA Finetuner

Speech To Text

• STT - Speech To Text - 语音转文本
• ASR - Automatic Speech Recognition - 自动语音识别
• whisper
• FunASR
  • MIT, Python
  • by 阿里
• wenet-e2e/wenet
  • Apache-2.0, Python, C++
  • End-to-End Speech Recognition Toolkit
• SpeechBrain
• Kaldi
• jianfch/stable-ts
  • MIT, Python
• Universal Speech Model
  • https://news.ycombinator.com/item?id=35365399
• ggerganov/whisper.cpp
  • MIT, C/C++
  • Port of OpenAI's Whisper model in C/C++
• chidiwilliams/buzz
  • MIT, Python
  • 桌面应用
  • transcribes and translates audio offline
• SYSTRAN/faster-whisper
  • MIT, Python
  • Whisper transcription with CTranslate2
• m-bain/whisperX
  • BSD-4, Python
  • Word-level Timestamps
• FL33TW00D/whisper-turbo
• collabora/WhisperSpeech
• Xenova/distil-whisper-web
• distil-whisper/distil-large-v2
• MahmoudAshraf97/whisper-diarization
  • Speaker Diarization based on OpenAI Whisper
• oTranscribe
  • https://otranscribe.com/
  • https://otranscribe.com/opensource/
• Speaker Diarization
  • 分离说话者
• 商业
  • https://spectropic.ai/
  • https://audiogest.app/

Text to Audio

• suno-ai/bark
  • https://news.ycombinator.com/item?id=35643219

Text to Music

• https://aitestkitchen.withgoogle.com/experiments/music-lm

Transfomer

• lm-sys/FastChat
  • platform for training, serving, and evaluating LLAMA-based chatbots
• jaymody/picoGPT
  • GPT-2
• hyperonym/basaran
  • 针对 HF Transformer 提供 OpenAI 兼容接口
• HarderThenHarder/transformers_tasks
• allenai/RL4LMs
• lucidrains/PaLM-rlhf-pytorch

Bot

• ConnectAI-E/Feishu-OpenAI
  • GPLv3, Go
  • /webhook/event
  • /webhook/card
  • https://open.feishu.cn/app/
  • 订阅 机器人进群, 接收消息, 消息已读
  • 权限
    • im:resource 图片 - 获取与上传图片或文件资源
    • im:message - group_at_msg,group_at_msg:readonly,p2p_msg,p2p_msg:readonly,send_as_bot
    • im:chat,im:chat:readonly
• https://github.com/zhayujie/bot-on-anything
• chathub-dev/chathub
  • All-in-one chatbot client

Misc

• huggingface/optimum
  • 加速、优化
• skypilot-org/skypilot
  • 在 cloud 上运行 ml
• CarperAI/trlx
  • 分布式训练
  • RLHF
• https://streamlit.io/

框架

• microsoft/onnxruntime
  • ONNX Runtime
• hidet-org/hidet
• mojo
• https://mlc.ai/web-llm/
  • https://news.ycombinator.com/item?id=35583349
• NodeJS
  • erelsgl/limdu
  • https://github.com/a16z-infra/ai-getting-started

Awesome

• webpilot-ai/Webpilot
  • GPLv3
• outlines-dev/outlines
  • Apache-2.0

Client

• Web
  • vercel/ai
    • https://sdk.vercel.ai/docs

Model

Model Awesome

Train/Alternative/TBD

• https://github.com/lm-sys/FastChat
• https://github.com/lvwzhen/law-cn-ai
• https://github.com/okisdev/ChatChat
• https://github.com/closedai-project/closedai
• https://github.com/openlm-research/open_llama
• https://github.com/ixaxaar/VardaGPT
  • https://news.ycombinator.com/item?id=35839536

Languages

• Stability-AI/StableLM
  • HN

3D

• Action-GPT: Leveraging Large-scale Language Models for Improved and Generalized Action Generation
  • https://actiongpt.github.io/
• openai/shap-e

Visual

• facebookresearch/dinov2

Image

Video

• nadermx/backgroundremover

Pricing

阿里云 语音交互

资源包

Item	Spec	Price	Unit Price
录音文件识别	40 小时	100 元	2.50 元/小时
	1,000 小时	1,200 元	1.20 元/小时
	20,000 小时	20,000 元	1.00 元/小时
	100,000 小时	90,000 元	0.90 元/小时
	250,000 小时	200,000 元	0.80 元/小时

后付费

item	标准价	每日用量	梯度折扣价
实时语音识别	3.50元/小时	0～299小时	3.50元/小时
		300～999小时	3.00元/小时
		1000～2999小时	2.40元/小时
		3000～4999小时	1.60元/小时
		5000 +	1.20元/小时

• 梯度折扣价
  • Tiered Discount Pricing
  • Graduated Discount Pricing
• https://help.aliyun.com/zh/isi/product-overview/billing-10

Learning

Reading, Learning, Tutorials, 教程, 阅读资料

• https://github.com/NielsRogge/Transformers-Tutorials