update

Author: wenerme

notes/ai/llm/llm-awesome.md

@@ -115,6 +115,8 @@ tags:
   - [coder-hxl/x-crawl](https://github.com/coder-hxl/x-crawl)
     - MIT, TS
     - Node.js AI-assisted crawler library
+  - [andrewyng/translation-agent](https://github.com/andrewyng/translation-agent)
+    - Agentic translation using reflection workflow
 - Coding
   - [replit/ReplitLM](https://github.com/replit/ReplitLM)
   - [albertan017/LLM4Decompile](https://github.com/albertan017/LLM4Decompile)
@@ -137,6 +139,16 @@ tags:
   - [llm-attacks/llm-attacks](https://github.com/llm-attacks/llm-attacks)
   - https://huggingface.co/blog/mlabonne/abliteration
     - [HN](https://news.ycombinator.com/item?id=40665721)
+- Promopt
+  - [langgptai/LangGPT](https://github.com/langgptai/LangGPT)
+    - 结构化提示词
+  - [linexjlin/GPTs](https://github.com/linexjlin/GPTs)
+    - leaked prompts of GPTs
+- Vision
+  - [vikhyat/moondream](https://github.com/vikhyat/moondream)
+    - tiny vision language model
+- [stanfordnlp/dspy](https://github.com/stanfordnlp/dspy)
+  - DSPy: The framework for programming—not prompting—foundation models
 - [BloopAI/bloop](https://github.com/BloopAI/bloop)
   - Answer questions about your code with an LLM agent
 - [HazyResearch/flash-attention](https://github.com/HazyResearch/flash-attention)

notes/ai/llm/llm-faq.md

@@ -16,23 +16,47 @@ tags:
 
 ## Pricing
 
-|                                 Model | 1M input/output |
-| ------------------------------------: | --------------: |
-|                                gpt-4o |  $5.00 / $15.00 |
-|                           gpt-4o-mini |  $0.150 / $0.60 |
-|                            o1-preview | $15.00 / $60.00 |
-|                               o1-mini |  $3.00 / $12.00 |
-|               Anthropic Claude 3 Opus | $15.00 / $75.00 |
-|     Groq Llama 3.1 70B Versatile 128k |   $0.59 / $0.79 |
-|                 Groq Whisper V3 Large |        $0.111/h |
-| DeepInfra Llama-3.1-70B-Instruct 128k |   $0.35 / $0.40 |
-|               DeepInfra Qwen2-72b 32k |   $0.35 / $0.40 |
-
-
+|                                 Model |  1M input/output | notes     |
+| ------------------------------------: | ---------------: | --------- |
+|                                gpt-4o |   $5.00 / $15.00 |
+|                            o1-preview |  $15.00 / $60.00 |
+|                               o1-mini |   $3.00 / $12.00 |
+|                           gpt-4o-mini |   $0.15 / $00.60 |
+|                   Gemini 1.5 Pro 128K | $3.5.00 / $10.50 |
+|                 Gemini 1.5 Flash 128K |  $0.075 / $00.30 |
+|               Gemini 1.5 Flash > 128K |   $0.15 / $00.60 |
+|            Gemini 1.5 {Flash/Pro}-002 |              50% | limits\*2 |
+|                Claude 3.5 Sonnet 200K |   $3.00 / $15.00 |
+|                   Claude 3 Haiku 200K |   $0.25 / $01.25 |
+|               Anthropic Claude 3 Opus |  $15.00 / $75.00 |
+|     Groq Llama 3.1 70B Versatile 128k |   $0.59 / $00.79 |
+|                 Groq Whisper V3 Large |         $0.111/h |
+| DeepInfra Llama-3.1-70B-Instruct 128k |    $0.35 / $0.40 |
+|               DeepInfra Qwen2-72b 32k |    $0.35 / $0.40 |
+
+:::tip
+
+- 小模型里 gpt-4o-mini 目前是性价比最好的
+- 开源模型的速度可以非常快，能实现 multi agent 这种模式
+- Gemini 1.5 Flash 支持 1M context window
+- Gemini 1.5 Pro 支持 2M context window
+
+:::
+
+| app          | price              | quota             |
+| ------------ | ------------------ | ----------------- |
+| ChatGPT Plus | $20                | 4o 80/3h, 4 40/3h |
+| ChatGPT Team | $25/年付, $30/月付 | 2\*Plus           |
+
+- ChatGPT Plus/Team 限制
+  - https://help.openai.com/en/articles/6950777-what-is-chatgpt-plus#h_d78bb59065
 - https://www.together.ai/pricing
 - https://deepinfra.com/pricing
 - https://groq.com/pricing/
 - https://openai.com/api/pricing/
+- https://www.anthropic.com/pricing
+- 参考
+  - https://www.vellum.ai/blog/llama-3-1-70b-vs-gpt-4o-vs-claude-3-5-sonnet
 
 ## model metrics
 

notes/ai/ml/cvat.md

@@ -6,6 +6,7 @@ title: CVAT
 
 - [cvat-ai/cvat](https://github.com/cvat-ai/cvat)
   - MIT, Python, TS
+  - Powerful and efficient Computer Vision Annotation Tool (CVAT)
   - by Intel
   - 支持 OpenCV
   - 支持基于模型的自动化标注

notes/ai/ml/gan.md

@@ -97,3 +97,5 @@ title: GAN
 - [From Scratch - GAN](https://ym2132.github.io/GenerativeAdversarialNetworks_Goodfellow)
   - https://github.com/YM2132/YMPaperImplementations/blob/main/paper_implementations/python_implemenations/GAN_Goodfellow.py
 - https://poloclub.github.io/ganlab/
+- [hindupuravinash/the-gan-zoo](https://github.com/hindupuravinash/the-gan-zoo)
+  - A list of all named GANs!

notes/ai/ml/jax.md

@@ -5,6 +5,10 @@ title: JAX
 # JAX
 
 - [google/jax](https://github.com/google/jax)
+  - Apache-2.0, Python, C++
+  - Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
 - 参考
   - [vopani/jaxton](https://github.com/vopani/jaxton)
     - exercises to learn JAX
+  - [felafax/felafax](https://github.com/felafax/felafax)
+  - [PyTorch is dead. Long live JAX.](https://neel04.github.io/my-website/blog/pytorch_rant/)

notes/ai/ml/libs/ml-libs-awesome.md

@@ -31,5 +31,4 @@ tags:
 - [flash_attn](https://github.com/Dao-AILab/flash-attention) - Flash Attention
   - 不支持 macOS/Apple Silicon
 - timm - PyTorch Image Models
-- [ml-tooling/best-of-ml-python](https://github.com/ml-tooling/best-of-ml-python)
 - https://pypi.org/classifiers/

notes/ai/ml/ml-awesome.md

@@ -116,6 +116,9 @@ tags:
     - https://magnific.ai/
 - image
   - https://huggingface.co/datasets/vivym/midjourney-messages
+  - [openai/CLIP](https://github.com/openai/CLIP)
+    - MIT
+    - CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
 - video
   - [blakeblackshear/frigate](https://github.com/blakeblackshear/frigate)
     - MIT, Python, TS
@@ -135,9 +138,8 @@ tags:
     - XCiT
     - DINO - Self-Supervised Vision Transformers
     - PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO
-- text & image
-  - [Yutong-Zhou-cv/Awesome-Text-to-Image](https://github.com/Yutong-Zhou-cv/Awesome-Text-to-Image)
-- tts
+- audio/music/speech/voice/tts
+  - [microsoft/muzic](https://github.com/microsoft/muzic)
   - [yl4579/StyleTTS2](https://github.com/yl4579/StyleTTS2)
     - MIT, Python
     - [HN](https://news.ycombinator.com/item?id=38335255)
@@ -188,6 +190,9 @@ tags:
     - NVIDIA Research
   - [THUDM](https://github.com/THUDM)
     - 清华 KEG & 数据挖掘
+- apps/ios/android
+  - [niedev/RTranslator](https://github.com/niedev/RTranslator)
+    - Apache-2.0, C++, Java
 - [PKU-YuanGroup/Video-LLaVA](https://github.com/PKU-YuanGroup/Video-LLaVA)
 - [GokuMohandas/MadeWithML](https://github.com/GokuMohandas/MadeWithML)
 - [josephmisiti/awesome-machine-learning](https://github.com/josephmisiti/awesome-machine-learning)
@@ -342,9 +347,23 @@ tags:
 
 ---
 
+- [ml-tooling/best-of-ml-python](https://github.com/ml-tooling/best-of-ml-python)
 - [TheAlgorithms/Python](https://github.com/TheAlgorithms/Python)
 - [freeCodeCamp/freeCodeCamp](https://github.com/freeCodeCamp/freeCodeCamp)
 
+## UI/Desktop/GUI/WebUI {#ui}
+
+- [lobehub/lobe-chat](https://github.com/lobehub/lobe-chat)
+  - Apache-2.0, TypeScript
+- [JetBrains/KotlinDL](https://github.com/JetBrains/KotlinDL)
+  - Kotlin DSL for ML
+
+## Library
+
+- js/ts/nodejs
+  - [dexaai/openai-fetch](https://github.com/dexaai/openai-fetch)
+  - vercel/ai
+
 ## Learn
 
 - [google-research/google-research](https://github.com/google-research/google-research)
@@ -485,16 +504,6 @@ tags:
 - [PyTorch vs TensorFlow in 2022](https://www.assemblyai.com/blog/pytorch-vs-tensorflow-in-2022/)
   - [HN](https://news.ycombinator.com/item?id=29552665)
 
-## Service
-
-- [openvinotoolkit/cvat](https://github.com/openvinotoolkit/cvat)
-  - Powerful and efficient Computer Vision Annotation Tool (CVAT)
-
-## Language
-
-- [JetBrains/KotlinDL](https://github.com/JetBrains/KotlinDL)
-  - Kotlin DSL for ML
-
 ## Intrested
 
 - [ZHKKKe/MODNet](https://github.com/ZHKKKe/MODNet)
@@ -521,10 +530,6 @@ tags:
 - https://modelplace.ai/models
 - [OpenBMB/BMList](https://github.com/OpenBMB/BMList)
 
-## Music
-
-- [microsoft/muzic](https://github.com/microsoft/muzic)
-
 ## STT
 
 - [snakers4/silero-models](https://github.com/snakers4/silero-models)

notes/ai/ml/stable-diffusion/README.md

@@ -13,6 +13,16 @@ title: Stable Diffusion
 - SD Turbo
 - [stabilityai/stable-video-diffusion-img2vid-xt](https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt)
   - Image-to-Video
+- [vladmandic/automatic](https://github.com/vladmandic/automatic)
+  - AGPLv3, Python
+  - SD.Next: Advanced Implementation of Stable Diffusion and other Diffusion-based generative image models
+- [ChenWu98/cycle-diffusion](https://github.com/ChenWu98/cycle-diffusion)
+- studio/webui
+  - [modelscope/DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
+- models
+  - https://huggingface.co/cagliostrolab/animagine-xl-3.1
+- 参考
+  - https://stable-diffusion-art.com/
 
 ---
 

notes/algorithm/cv/imagemagick.md

@@ -5,6 +5,7 @@ title: ImageMagick
 # imagemagick
 
 
+- [ImageMagick/ImageMagick](https://github.com/ImageMagick/ImageMagick)
 - http://www.imagemagick.org/
 - http://www.imagemagick.org/script/formats.php
   - PNG, JPEG, GIF, HEIC, TIFF, DPX, EXR, WebP, Postscript, PDF, and SVG
@@ -43,3 +44,14 @@ pdftk pdf1.pdf pdf2.pdf cat output temp.pdf
 
 magick mogrify -monitor -format jpg *.HEIC
 ```
+
+# FAQ
+
+## The convert command is deprecated in IMv7, use "magick" instead of "convert" or "magick convert"
+
+- `convert` -> `magick convert`
+- IM7
+
+---
+
+- convert, mogrify, identify, composite -> magick

notes/blockchain/eth/chainlist.md

@@ -15,6 +15,11 @@ Network 列表，使用 ID 标识，每个 Network 有自己的 本地货币/Nat
   - Explorer https://eips.ethereum.org/EIPS/eip-3091
   - 模型参考 https://github.com/ethereum-lists/chains/blob/master/_data/chains/eip155-1.json
   - https://github.com/satoshilabs/slips/blob/master/slip-0044.md
+  - https://tron.network/
+    - tron20
+    - TRONIX mainnet native token
+    - 波场
+    - DPoS（Delegated Proof of Stake，代理权益证明）
 
 | Chain                       | Symbol | Date    | Proof |
 | --------------------------- | ------ | ------- | ----- |