AI Code and Productivity Tools
Local Large Language Models (LLM) Tools
Section titled “Local Large Language Models (LLM) Tools”Sources:
- Forget ChatGPT: why researchers now run small AIs on their laptops - advantages of local AI
- 5 easy ways to run an LLM locally - Details on options and their software features
Advantages of local AI:
- Consistency
- Works offline and not tied to service changes
- Privacy
- Cost
Software for local AI
- Ollama (available for Mac, Windows and Linux
operating systems) lets users download open models, including Llama
3.1, Phi-3, Mistral and Gemma 2, and access them through a command
line
- Model downloads are easy
- Integration with LangChain, PrivateGPT
- GPT4All cross platform app
- Chat like GUI
- Can chat with own local documents
- Download models
- Integration with CLI, local server API like OpenAI API, Python, Node
- Llamafile, which can
transform LLMs into a single file that runs on several operating
systems, with or without a graphics processing unit
- Single executable LLM files
- Supports GGUF format for import and run
- Chat in browser
- LLM by Simon Willison
- CLI, requires Python installed
- CLI customization
- Uses OpenAI models and reuse gpt4all ones
- CLI, requires Python installed
- h2oGPT - good for advanced users
- Chat GUI
- Local documents chat
- Model configuration
- PrivateGPT /
LocalGPT - query own
local documents
- “PrivateGPT features scripts to ingest data files, split them into chunks, create “embeddings” (numerical representations of the meaning of the text), and store those embeddings in a local Chroma vector store. When you ask a question, the app searches for relevant documents and sends just those to the LLM to generate an answer.”
- Requires Python
Additional Options
Section titled “Additional Options”- Jan - Model download, RAG
- Opera browser (developer stream) - local AI option
- NVIDIA Chat with RTX - Chat, RAG
- LM Studio
- Hugging Face as model source
- Good GUI, can use own data
- API same as OpenAI
- LangChain - framework for creating generative AI applications
Tools from Others
Section titled “Tools from Others”- Chat with PDFs - https://pensieve-app.springworks.in/
Examples of publicly available generative AI tools:
- Textual: Google’s Gemini, Open AI’s Chat GPT, Anthropic’s Claude
- Visual: DALL-E 3, Bing Image Creator, Midjourney, Photoleap, Adobe Firefly, Canva
- Audio: Boomy, Soundraw, Synthesia.io, Canva
Top 5 AI Tools other than ChatGPT 2023-04-25
Section titled “Top 5 AI Tools other than ChatGPT 2023-04-25”Source: Top 5 AI Tools That Are BETTER Than ChatGPT, But Nobody is Using Them | Coding & Productivity Tools - YouTube - Tiff in Tech
- AgentGPT
- Autonomous AI
- Goal based AI that does not require constant prompting
- Can make your own autonomous AI
- Ora.sh
- Build AI apps
- meetcody.ai
- Customized AI by feeding it data
- It will learn about your data like business data, trained with your data
- Then ask questions about your data
- Jasper
- Reword content in a certain tone of voice given your input
- literallyanything.io
- Builds entire application in short amount of time
10 AI Tools
Section titled “10 AI Tools”Source: 10 AI Tools YOU WON’T BELIEVE EXIST! - YouTube - Hayls World
- Lexica.art - art generator, like Mid Journey and Stable Diffusion
- Generate from prompt and optional input images
- Descript - voice cloner, record your voice, transcribe, train AI, AI can generate your voice speaking from text
- Mubert - Music generated by Mubert https://mubert.com/render, AI generative music, choose genres
- Uberduck.ai - text to speech lyrics to music, choose voice like singers
- Nvidia Canvas - painter, brushes are things like hills, cloud, etc. Paint in a blue screen and it will generate on top of photo, can choose designs
- Replicate.com - photo restoration, colouring and other photo manipulation
- Lumalabs.ai - video recording of object to 3D, AI scans and produces 3D model, allows examination of 3D model
- ChatGPT - chat with AI, plugins, Chrome extensions:
- See below
- Supermeme.ai - meme generator
- Kaiber.ai - start with image, video generator based on prompt, choose style, camera settings, changes in video
Use cases for Generalized AI
Section titled “Use cases for Generalized AI”- Summary with GPT - summary YouTube video you are watching
- AIPRM for ChatGPT - prompt templates to generate things like articles, documents
- 10 ChatGPT Life Hacks -
YouTube Hayls World -
use cases for ChatGPT:
- Summarise:
- Articles, YouTube videos with a link
- For YouTube ask it to summarize and provide the title and video title
- Text: “TLDR in 2 sentences” <paste text>
- e.g. legal documents, agreements, documents
- Images from URLs, with metadata if necessary
- Articles, YouTube videos with a link
- Similar to summarize, parse content and write:
- e.g. job description to cover letter
- Learning - teach / Grade papers
- Teach me… to get a guide
- e.g. software, languages, music
- Teach me… to get a guide
- Health
- Ask regarding exercise plan
- Meal plans
- Calories allowances
- Shopping lists for diets
- Recipes
- What can be cooked with prompt inputs
- Multi-identity
- Ask it to analyze your messages and write like you or like someone else
- Ask AI to pretend to be someone else (person, type of person) and respond like them
- Write responses/dealing with tough situations
- Leisure:
- Suggest media
- Play games (ask what games it knows)
- e.g. trivia
- Code games in browser
- Tips:
- Assign persona to GPT to response, some respond better
- Correct ChatGPT
- Summarise: