When you become a Patron you gain access to my courses along with the code to all my projects
1000x Dev MasterClass
Streamlit Course
FastAPI Course
echohive AI Academy
400+ Free coding videos
Learn to code AI powered Apps with my free videos! Become a Patron to download project files and gain access to my courses and 1-on-1 meetings
About echohive AI Academy
Hi. Thank you for visiting my projects. I am a self-taught full stack AI developer. I have created this platform to share my knowledge with the world. I hope you find the content useful. Please feel free to reach out to me on Twitter or Discord.
Sign Up for Updates
1000x MasterClass
In this MasterClass I will be sharing everything I learned on how to use AI assistance for coding as effectively as possible to become 1000x coders eventually 🚀
As many of you know I have put in over 3000 hours to create 300 projects in the last year and a half and I would like to share what I have learned so we can all get better and build better 💪
I feel this is the most important and useful piece of the puzzle in coding. I will be updating this MasterClass regularly.
Veo-2 auto short movie generator in python
Download the code: https://www.patreon.com/posts/veo-2-auto-story-126307938
Join my AI community: https://www.patreon.com/c/echohive42
Watch the full video exploring 8 Cursor agentic use cases: https://www.patreon.com/posts/convert-cursor-8-126098688
Join my AI community: https://www.patreon.com/c/echohive42
Controlling & Directing Claude Sonnet's thinking process
Download the code and watch the full tutorial: https://www.patreon.com/posts/influencing-for-125398339
Join my AI community: https://www.patreon.com/c/echohive42
How to build apps with thousands of lines of code in Cursor
Watch the full process + code: https://www.patreon.com/posts/3-23-1000x-lab-125006367?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Join my AI community: https://www.patreon.com/c/echohive42
Watch the full video and download the code: https://www.patreon.com/posts/3-13-1000x-lab-124331718
My 1000x Cursor course: https://www.patreon.com/collection/520959
Testing and building with Computer Use from OpenAI
Download the code and watch the full video: https://www.patreon.com/posts/3-12-1000x-lab-124186443
My 1000x Cursor course: https://www.patreon.com/collection/520959
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
Merging Minds of DeepSeek-r1, 2.0-Flash-thinking, QWQ, and o1 for Ultimate Answers! Open Source💝
Watch how I built this step by step: https://www.patreon.com/posts/building-chat-o1-120572049
Repo link: https://github.com/echohive42/frankenthought-chat
Cursor pretends to be o1 and improves itself in any way it wants to become
This is part of my 1000x Course. join by becoming a patron and unlock Cursor superpowers!
My 1000x Cursor course: https://www.patreon.com/collection/520959
Watch the full video: https://www.patreon.com/posts/using-cursor-as-120158280
View benefits of different Patreon memberships: https://www.patreon.com/c/echohive42/membership
How to create agent systems with OpenAI Swarm library | Open Source 💝
Github link to all files: https://github.com/echohive42/Swarm-library-tutorials
Watch how I built this from scratch + 30 other projects: https://www.patreon.com/posts/how-to-build-119979655
My AI Cracked the Top 1% on Kaggle🥳 – So I open Sourced it 💝
Watch and learn how I built this: https://www.patreon.com/posts/updated-how-to-1-112197565
Download it from Github:https://github.com/echohive42/AI-Data-Scientist-scores-top-1-percent-on-Kaggle
How to build a Form Filling AI chat webapp | Full coding walkthrough
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
How to use "Rules For AI" for maximum speed in Cursor. Part of my 1000x course
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
chat with 1000 personal and business productivity agents. Try it online!
Try 1000 minds online: https://1000minds.up.railway.app/
Download all 1000 agent prompts and the full webapp code: (use "2DCDD" for 20% discount until Jan 3rd)
build code from diagram(trying to), practicing math, image guessing, ranking on chatbot arena
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
Gemini 2.0 FLASH Real-Time Screenshare video chat while testing its reasoning capabilities 🤯
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
"DONT SAY THIS" LLM game for promoting your business or products
Play the game here and win discounts and FREE membership to my patreon: https://dontsaythis.up.railway.app/
Download the game with one time purchase($100): https://www.patreon.com/echohive42/shop/dont-say-this-llm-game-for-promoting-or-770363
Real-time hand/face tracking + talk to Groq using hands and other gesture experiments
Download hand detection full stack starter template(FREE): https://www.patreon.com/posts/hand-gesture-you-117172734
Download 2 hands with 3d bars visualization(paid membership): https://www.patreon.com/posts/2-hand-tracking-117207206
Cursor creates its own tools and becomes an open ended agent
Download cursor rules: https://www.patreon.com/posts/cursor-rules-for-116995026
Download computer use cursor control with tools: https://www.patreon.com/posts/claude-computer-116852227
Fully autonomous cursor is being driven by Claude computer use + inception demo
Download computer use cursor control with tools: https://www.patreon.com/posts/claude-computer-116852227
Download computer use cursor control without tools:
Step by step build method with Cursor Composer and Windsurf Cascade. Cascade at the end of video
My 1000x Cursor course: https://www.patreon.com/collection/520959
View benefits of different Patreon memberships: https://www.patreon.com/c/echohive42/membership
My 1000x Cursor course: https://www.patreon.com/collection/520959
GET 20% Discount on Virtuoso membership until Nov 29: https://www.patreon.com/posts/join-as-virtuoso-116590490
Step by step way to use cursorrules file to build entire project descriptions iteratively in Cursor
My 1000x Cursor course: https://www.patreon.com/collection/520959
View benefits of different Patreon memberships: https://www.patreon.com/c/echohive42/membership
This app finds the best image for your content by polling gpt-4o-minis
Download the source code for this and 300+ other projects by becoming a Patron: https://www.patreon.com/posts/image-contest-b-116474918
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
LLM Stock predictions Monte Carlo simulator Python script
Download this project: https://www.patreon.com/posts/stock-price-with-116013401
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
Tinder-like swiping Arxiv paper finder Fast API Full stack webapp
The project is an innovative FastAPI app designed to connect researchers with relevant academic papers in a fun and engaging way. By utilizing a Tinder-like swiping interface, users can easily discover, match, and explore scholarly articles from Arxiv, making the process of finding research papers both intuitive and enjoyable.
Arxiv breakthrough synthesizer & new paper generator
This is a three-script pipeline that automates academic research paper synthesis. First, 1_arxiv_breakthrough_synthesizer.py searches arXiv for papers on a specific topic (like "reasoning in large language models") and uses AI to identify promising combinations of papers that could lead to breakthroughs. Then, 2_format_combinations.py takes these combinations and formats them into readable markdown and text files with proper links and formatting. Finally, 3_generate_papers.py generates entirely new research papers by combining the ideas from each paper combination, including downloading and analyzing the full text of the source papers from arXiv, and outputs detailed academic papers with standard sections like abstract, methodology, and conclusions.
using LLMs to predict stock prices simulation part 1
This tool is a multi-day stock price prediction analysis script using GPT-4o and historical stock data. It features parallel processing for efficiency, evaluates prediction accuracy, and calculates detailed error metrics (ME, MAE, RMSE). The tool provides direction accuracy analysis, visualizes predictions against actual prices, and generates comprehensive reports summarizing the results. It emphasizes educational use, cautioning against applying AI predictions as financial advice.
Download the source code for this and 300+ other projects by becoming a Patron: https://www.patreon.com/posts/new-openai-in-115380515
View benefits of different Patreon memberships: https://www.patreon.com/c/echohive42/membership
LLM Algorithm Battles: AI models try to predict the algorithm to produce a sequence of numbers
Download the source code for this and 350+ other projects by becoming a Patron: https://www.patreon.com/posts/algo-solver-llm-115688747
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
Realtime Voice without realtime API using Voice activity detection with groq whisper and openai
Download the source code for this and 300+ other projects by becoming a Patron: https://www.patreon.com/posts/realtime-voice-115556093
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
Analyzing Meta-Prompts: Assessing AI Summary Quality with OpenAI
Download the source code for this for free by becoming a free member at Patreon: https://www.patreon.com/posts/meta-prompts-ai-114965087
View benefits of different Patreon memberships: https://www.patreon.com/c/echohive42/membership
Auto Vision Test for Gpt-4o vs 3.5 Sonnet using Pygame in Python. who will win?
Download the source code for this and 300+ other projects by becoming a Patron: https://www.patreon.com/posts/iterative-vision-115371887
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
LLM assisted Form filling chat Resume Builder webapp using structured outputs
Watch how I built this in 30 minutes and download the source code by becoming a Patron: https://www.patreon.com/posts/28-how-to-build-115298942
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
Mind simulation with thoughts, memory retrieval, beliefs and conclusions formed
Download the source code for this and 300+ other projects by becoming a Patron: https://www.patreon.com/posts/llm-mind-version-115243967
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
I built an Stock Outlook Analysis bot: Your Own AI-Driven Stock Predictor with gpt-4o & FastAPI
Download the source code for this and 300+ other projects by becoming a Patron: https://www.patreon.com/posts/auto-stock-using-115115508
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
AI reads books(PDFs) page by page & synthesize knowledge using gpt-4o, o1-mini & structured outputs
Download the source code for this and 300+ other projects by becoming a Patron: https://www.patreon.com/posts/ai-reads-books-115101147
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
Download the source code for this and 300+ other projects by becoming a Patron: https://www.patreon.com/posts/automated-sales-115043908
Download the AI Code Explainer: https://www.patreon.com/posts/automated-ai-115037437
From AI Agents to Whisper: The Best of AI in Action – Demos & Tutorials! echohive best of!
Download 10+ Agents work together to build any Python app: https://www.patreon.com/posts/107301172
Download 8+ agent works together to develop games using Claude 3.5 Sonnet: https://www.patreon.com/posts/107301172
Translating long documents with GPT-4o + translation Eval | using structured outputs
Download the source code for this project: https://www.patreon.com/posts/translating-long-114015677
My 1000x Cursor course: https://www.patreon.com/collection/520959
Download the source code for this project: https://www.patreon.com/posts/real-time-voice-113657825
My 1000x Cursor course: https://www.patreon.com/collection/520959
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
Watch how to build o1 team manager and download the code: https://www.patreon.com/posts/9-15-architect-112125909
Download Groq writing assistant: https://www.patreon.com/posts/thoughts-to-text-112698689
Auto Prompt Engineer tests itself against 5 difficult problems
Download the Auto Prompt Engineer by becoming a Patron: https://www.patreon.com/posts/auto-prompt-5-111628684
My 1000x Cursor course: https://www.patreon.com/collection/520959
Transcribe hours long videos in minutes with Fast Transcriber
Download Fast Transcriber from my Patreon: https://www.patreon.com/posts/fast-transcriber-112321464
My 1000x Cursor course: https://www.patreon.com/collection/520959
ElevenLabs: Easily Replace Video Audio with AI Voices Using Python!
Download this project files: https://www.patreon.com/posts/video-voice-over-111569263
My 1000x Cursor course: https://www.patreon.com/collection/520959
Full Stack Auto Coder builds FastAPI webapps using o1-mini and Claude Sonnet
Download the project and watch How I built this: https://www.patreon.com/posts/how-to-build-o1-112010322
My 1000x Cursor course: https://www.patreon.com/collection/520959
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
Cursor codes Single file full stack ChatGpt clone in 1 minute in ~100 lines
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
My 1000x Cursor course: https://www.patreon.com/collection/520959
watch the first chapter for FREE: https://www.patreon.com/posts/cursor-deep-dive-110897542
Join my 1000x Cursor course by becoming a Patron: https://www.patreon.com/posts/1000x-code-with-103326330
Download this project by becoming a Patron: https://www.patreon.com/posts/exclusive-bonus-110721456
Coding in Cursor: 3 full stack webapps built in 5 minutes. Chatgpt clone, Voice notes, landing page
Join my 1000x Cursor course by becoming a Patron: https://www.patreon.com/posts/1000x-code-with-103326330
Download this project by becoming a Patron: https://www.patreon.com/posts/exclusive-bonus-110721456
Coding with Cursor in 1 MINUTE! Building a Full stack Video summarizer
Join my 1000x Cursor course by becoming a Patron: https://www.patreon.com/posts/1000x-code-with-103326330
Downlead this project by becoming a Patron: https://www.patreon.com/posts/exclusive-bonus-110721456
Evaluation system for Creativity Using GPT-4 with JSON mode and Structured Outputs
Download Ideation LLM Alchemies at my Patreon along with 250+ other projects:https://www.patreon.com/posts/evaluation-for-110635954
Echohive Hackathon: https://www.patreon.com/posts/echohive-18-31-110344393
Download AI SCIENTIST at my Patreon along with 250+ other projects:https://www.patreon.com/posts/how-to-build-ai-110553257
Echohive Hackathon: https://www.patreon.com/posts/echohive-18-31-110344393
Download this project files including winner apps folder at my Patreon along with 250+ other projects:https://www.patreon.com/posts/fine-tune-gpt-4o-110489244
Download this project without the winner apps folder: https://www.patreon.com/posts/fine-tune-gpt-4o-110489239
Synthetic Thought Frameworks for Novel Code generation
Download Synthetic Thought Frameworks and 400+ generated apps at my Patreon along with 250+ other projects:https://www.patreon.com/posts/synthetic-with-110364361
Download just the Synthetic Thought Frameworks without 400+ generated apps:https://www.patreon.com/posts/synthetic-400-110364367
Save %90 and reduce latency with Claude prompt caching
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/save-90-and-with-110133380
Download the basic files for FREE: https://www.patreon.com/posts/claude-prompt-by-110106313
Lightning Fast Builds: 6 AI Apps in 15 Minutes Using Cursor IDE!
Download this Project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/lightning-fast-6-110067625
Talk with me this Wednesday the 14th, AMA for Architect+ level Patrons: https://www.patreon.com/posts/ama-meetings-for-109136215
Live genetic Algorithm coding with Cursor. AI learns to play snake
Download the project files at my 1000x MasterClass + 15 other project videos: https://www.patreon.com/posts/1000x-dev-103326330\
Learn to code fast with AI assistance with my 1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330\
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/1000x-dev-103326330\
Learn to code fast with AI assistance with my 1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330\
Add memories to your AI chat apps
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/add-memories-to-109823010
Generate Web Components with The new Structured outputs from OpenAI
In this video we will use the new gpt-4o-2024-08-06 with structured outputs to create a webapp which generates daisy ui web components and an iterative autocoder.
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/generate-web-new-109662046
Building a ChatGPT clone with gpt-4o, claude 3.5 sonnet, gemini 1.5 pro using Cursor Composer only
Building a ChatGPT clone with gpt-4o, claude 3.5 sonnet, gemini 1.5 pro using Cursor Composer only
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/building-chatgpt-109560852
Extracting Knowledge Graphs and Structured Data from very long PDF files
Extracting Knowledge Graphs and Structured Data from very long PDF files using gpt-4o-mini and gpt-4o.
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/extracting-and-109421779
Instant Interactive Globe with city explorer using Groq API
Groq Globe explorer
Download the All Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/globe-explorer-109288620
6 Agent customer service team with function calling gpt-4o
This agent network consists of 6 agents, one manager customer service agent which routes to all other agents.
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/how-to-build-109234353
Streamlit Knowledge Graph: Unearth Insights from PDFs and Text Files
create kowledge graphs from text or PDF files and create sumamries and json entity extraction.
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/streamlit-graph-109131557
Chat with an entire library of 100 books and 17 million tokens using RAG, Chromadb and GPT-4o.
Chat with a Whole Library using RAG, Chromadb and GPT-4o
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/chat-with-whole-108904494
Step-by-Step Guide: Fine-Tuning GPT-4o Mini with Synthetic Data.Dataset can be created manually also
This script generates synthetic data to fine tune gpt-4o-mini. You can also use the script to generate your own custom dataset.
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/step-by-step-gpt-108838646
Testing Llama 3.1 reasoning with 5 increasingly difficult problems
Testing Llama 3.1 reasoning with 5 increasingly difficult problems
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/testing-llama-3-108759331
This is a preview from my 1000x MasterClass. Full course videos and project files at the link below.
Learn to code fast with my 1000x MasterClass:https://www.patreon.com/posts/1000x-dev-23rd-3-103326330
Self Instructing Agent Chatbot writes its own system messages
Self Instructing Agent Chatbot writes its own system messages. It can also save and load system messages
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/self-instructing-108628219
testing GPT-4o MİNİ for reasoning, structured data from receipts and product images vision tests
GPT-4o Mini reasoning, structured data from receipt images and product images vision tests
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/gpt-4o-mini-data-108414519
Agent AGI with Task Scheduler and a much improved reasoning prompt for tool usage
This agent which uses Anthropic's Claude 3.5 Sonnet begins with a single tool, a tool which can create other tools. Then builds its own tools. sky is the limit
Download AGENT AGI at my Patreon along with 250+ other projects:https://www.patreon.com/posts/agent-agi-builds-107481847
This video showcases an innovative AI-powered coding assistant called CoderTeam. It leverages multiple language models like Claude, GPT-4o, Gemini, and DeepSeek to collaboratively discuss, generate, and refine code projects. The system guides users through project description, team discussion, initial code generation, error correction, and iterative improvements based on user feedback.
Unified LLM APIs, makes it super easy to build any LLM-Agent app
This python class combines OpenAI, Anthropic and Openrouter apis for suepr easy LLM implementations. default mdoels are GPT-4o, Claude 3.5 Sonnet and Gemini 1.5 pro but you can change the models to anything you like. requires openrouter, openai, antrhopic api keys for those models.
Download the full Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/unified-llm-apis-108090093
8+ Agents work together to author a book + audiobook + book webpage
8+ Agents work together to author a book
Download Agentic book maker at my Patreon along with 250+ other projects:https://www.patreon.com/posts/8-agents-work-to-108040259
Agent AGI UPDATED with new voice control and improvements + Code review.
Agent AGI is a python LLM agent built with Anthropic's Claude 3.5 Sonnet.it now has vision and you can instruct it with your voice as well.
Download Agent AGI at my Patreon along with 250+ other projects:https://www.patreon.com/posts/agent-agi-writes-107481847
In this video we will explore a constrained agentic system which can spawn over 10 agents to build any python program using Anthropic's Claude 3.5 Sonnet. there are 2 planning agents, 1 error correcting agent, 1 feedback agent and many file writer agents
Download Agentic Dev Team at my Patreon along with 250+ other projects:https://www.patreon.com/posts/10-agents-work-107799448
LLM routing explained with 3 examples, simple to advanced
LLM routing explained with 3 examples, simple to advanced
Download Project files for this at my Patreon along with 250+ other projects:https://www.patreon.com/posts/llm-routing-with-107725033
Agent AGI gains Vision and becomes a Financial Analyst by writing its own tools
Agent AGI is a python LLM agent built with Anthropic's Claude 3.5 Sonnet.it now has vision and you can instruct it with your voice as well.
Download Agent AGI at my Patreon along with 250+ other projects:https://www.patreon.com/posts/agent-agi-writes-107481847
Agent AGI builds its own tools, takes multiple actions, improves itself
This agent which uses Anthropic's Claude 3.5 Sonnet begins with a single tool, a tool which can create other tools. Then builds its own tools. sky is the limit
Download the Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/agent-agi-writes-107481847
8+ agents work together to develop games using Claude 3.5 Sonnet
In this project we will explore a semi autonomous agentic workflow which starts up agents to build a multi-file game in pygame using Anthropic's Claude 3.5 Sonnet
Download the Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/8-agent-works-to-107301172
Muti Panel LLM chat with Next JS, Vercel AI sdk. with separate system messages and model selection
Muti Panel LLM chat with Next JS, Vercel AI sdk. with separate system messages and model selection
Download the Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/muti-panel-llm-107146246
Deep RL Agent learns to play snake game from scratch
Deep RL Agent learns to play snake game from scratch
Download the files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/deep-rl-agent-to-107075051
Next JS: Transcribe and rewrite with Groq API or GPT-4o
Transcribe and rewrite with Groq API or GPT-4o
Download the Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/transcribe-and-106836038
Ultimate Web Researcher with Exa & GPT-4o & reranking algo
this web researcher uses a combination of gpt and exa api calls along with custom chunking and cosine similarity reranking algorithm to try to get the best answers for each user questions
Download the Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/gpt-4o-exa-multi-106754304
Tool/Function use deep-dive with Claude 3.5 Sonnet explained A to Z with 10 examples
Function calling deep-dive with Claude 3.5 Sonnet explained A to Z with 10 examples
Download the Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/tool-function-3-106656952
Testing Claude 3.5 Sonnet | reasoning, complex image chart, building a tower defense game
In this video we will be testing the new 3.5 Sonnet model from Anthropic with a very complex image chart, build a tower defense game with it using 8-bit art, test it against 3 difficult reasoning problems
Download ALL files for this project at my patreon: https://www.patreon.com/posts/all-files-for-3-106578697
Modular Rag and chat implementation from URLs, PDFs and txt files.
This project has many modular pure python functions to split text with overlap, embed text and cosine similarity and more so you can experiment and try to come up with your own rag strategies.
Download the Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/modular-rag-and-106461497
Cursor IDE deep dive and why it is the best tool for coding
we will explore many of the awesome features of cursor, such as cursor chat, ctrl+k command, long context and interpreter modes, as well as documentation vectorizer, definitions command and .cursorrules file
Learn to code fast with my 1000x MasterClass:
Multiple LLM Agents perform analysis on an ongoing conversation. Next JS project with Vercel AI SDK
In this Next JS project we can have up to 8 agents each with a custom system message which can perform any type of analysis on the ongoing conversation based on default system messages or from your custom instructions.
Download the Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/multiple-llm-on-106220356
Easy web Scraping, crawling, web search for LLMs with GPT-4o chat
Easy web Scraping + Crawl + Search with Firecrawl including chat with GPT-4o using the context
ALL Project files for this project at my Patreon along with 250+ other projects:https://www.patreon.com/posts/all-files-for-4o-106035051
Easy Web scraping + search with Jina AI extracts clean text for LLM context including PDF urls
Jina AI which you can use with or without an API key which returns a clean text from urls or web search.
Project files for this project at my Patreon along with 250+ other projects https://www.patreon.com/posts/easy-free-web-ai-105970057
Multi Modal Rag system with chromadb using only opensource embeddings
Multi Modal Rag system with chromadb using only opensource embeddings
Code files and coding walkthrough for this project available at my 1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330
GPT-4 Omni personas perform market research for you with real-time updating plot and coding walkthru
In this script we create "n" many personas based on a specific demographics and then ask these personas a particular question with given options. Then we plot them in real time and save the results to a json file
Learn to code fast with my 1000x MasterClass(code included for this project): https://www.patreon.com/posts/1000x-dev-103326330
Livekit real-time Ai agent voice interface in Python with Auto detected user interruptions
Livekit real-time Ai agent voice interface in Python with Auto detected user interruptions
Download Project files for this project at my Patreon: https://www.patreon.com/posts/livekit-real-ai-105674470
Next js + FastAPI + Python implementation of realtime chat sentiment analysis with GPT-4 Omni
This video features real time chat sentiment analysis in 3 implementations. Python terminal, Fastapi webapp and Next JS webapp
Download all Project files for this project including Next JS app at my Patreon: https://www.patreon.com/posts/all-files-for-js-105610496
Should you learn Next JS to build AI powered web applications?
Learn to code fast with my 1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330
How does Phi-3 mini, medium and Gemini 1.5 Pro and Flash stack up in reasoning?
How does Phi-3 mini, medium and Gemini 1.5 Pro and Flash stack up in reasoning?
Project files for this project at my Patreon along with 250+ other projects https://www.patreon.com/posts/how-does-phi-3-1-105341908
Creating a Dynamic Knowledge Distillation Web App Using FastAPI and GPT-4o
This is a fastapi webapp implementation of my previous knowledge distillation video which generates entire articles and markdown from gpt-4o generated sentences guided by the user given any topic
GPT-4o builds articles and markdown by first distilling knowledge sentence by sentence
In this script, user enters a topic then GPT-4 generates sentences. User can then expand on any sentence continuously. then user can decide to export the sentences as a text or markdown article
Project files for this project at my Patreon along with 250+ other projects https://www.patreon.com/posts/gpt-4o-builds-by-105177598
GPT-4 Omni creates DEEPER Knowledge trees based on any topic
GPT-4o creates Knowledge trees based on any topic.
Project files for this project at my Patreon along with 250+ other projects https://www.patreon.com/posts/gpt-4o-creates-105113958
GPT-4 Omni generates unique and detailed chat personas. includes saving and retrieving of personas
GPT-4 Omni generates personas fora chat. includes saving and retrieving of personas
Download project files and Watch the coding walkthrough video at my 1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330
GPT-4o auto generates html landing pages from Streamlit webapp
in this video we will take a streamlit webapp implementation which allows us to automatically generate custom html landing pages using OpenAI GPT4 Omni model
Download the files for this project at my Patreon along with 250+ other projects https://www.patreon.com/posts/gpt-4o-auto-html-104766020
GPT-4o creates Knowledge trees based on any topic.
Project files for this project at my Patreon along with 250+ other projects https://www.patreon.com/posts/gpt-4o-creates-104690966
TOP 15%!! GPT-4 Omni solves Space ship Titanic Kaggle data challenge
In this video we will use GPT4 in data analytics mode to try to solve A kaggle challenge "Spaceship Titanic"
1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330
Can GPT-4o watch and understand 10 videos at the same time?
we will get frames out of 10 videos, one per second, and send it all to gpt4 shuffled and text to see if it can understand them
Watch the coding walkthrough video at 1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330
Swarm Intelligence with GPT-4o, Gemini 1.5, Opus and Llama3 70b
Swarm Intelligence with GPT-4o, Gemini 1.5, Opus and Llama3 70b
Project files and coding walkthrough video at 1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330
This is a quick video on the capabilities of the new OpenAI model GPT-4o. truly a multi-modal model from ground up
Capabilities page: https://openai.com/index/hello-gpt-4o/
This is a simplified version of my Auto Streamer V3 without voice and with limited options
Project files at my Patreon: https://www.patreon.com/posts/auto-streamer-104090405
How to build multi threaded apps in streamlit with GPT API
We will explore how to build multi theraded apps which runs concurrent processes in streamlit using gpt 3.5 turbo with QA and summarization threads.
Download the project files at my Streamlit course: https://www.patreon.com/posts/exclusive-course-102751857
Testing good-gpt-2 and also-good-gpt2 in chatbot arena
This video is a series of experiments using good-gpt-2 and also-good-gpt2 chatbots in lmsys arena in battle mode
1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330
How to use Interpreter Mode and Terminal Commands in Cursor IDE so you can code FASTER!
We will explore how to use interpreter and terminal ctrl+K commands in cursor. Interpreter mode allows cursor to execute multiple actions automatically and terminal commands lets you write complex commands in terminal
1000x MasterClass: https://www.patreon.com/posts/1000x-dev-103326330
NOTE: use the models they arena(battle) not direct chat or side by side. They don’t seem to be available for those
free Code files for this project: https://www.patreon.com/posts/emergency-video-103788012
Hume two way voice chat with GPT summaries and saved message history
We will run a parallel threaded gpt sumarization on the realtime conversation we are having with Hume EVI, We will also save the conversation to a json file. We will modify the Hume library to achieve this
Exploring the new OpenAI Batch API in web and in Python code
Exploring the new OpenAI Batch API in web and in Python code
Free code files for this project: https://www.patreon.com/posts/exploring-new-in-103461572
Groq blazing fast Llama 3 70B gets instructed by GPT4
We can use Llama 3 with Groq API for super fast inference or Opus, Sonnet or Haiku from Anthropic's Claude API with this Auto coder. GPT4 handles the iterative instructions for automated coding
Hume AI Emotional Voice Interface(EVI) First look + Installation instructions for Python
In this video we will explore the Hume AI playground and run it from python as well
Code files for this project at Patreon: https://www.patreon.com/posts/first-look-at-ai-103192685
RAG Agents: How to Design AI Assistants with OpenAI API with 10.000 documents support
We will explore OpenAI Assistants API with a deeper look. We will see how we can create or retrieve Assistants from Python and upload files for a RAG system
Llama 3 70B Gamer Agent competes against Opus and GPT-4 in this Gamer Agents simulation Arena
we use a simple custom built Pygame Gamer Agent Simulation Arena to compare Meta's Llama 3 70b model to Claude 3 Opus and OpenAI's Gpt-4-turbo
Code files for this project at Patreon: https://www.patreon.com/posts/llama-3-70b-opus-102806589
Llama 3 70b vs 8b vs Mixtral 8x22b vs WizardLM 8x22b in a reasoning test
We test Llama 3 70b ,8b, Mixtral 8x22b, WizardLM 8x22b in a reasoning test using a shuffled combination of 2 difficult reasoning word/math problems
Code files for this project at Patreon: https://www.patreon.com/posts/llama-3-70b-vs-102634576
How to use Assistants API and the best way to do it
In this video we will explore on how we can use the Assistants web Ui to create out assistants and configure it then use it from our python script.
Code files for this project at Patreon: https://www.patreon.com/posts/explaining-how-102574497
Build any content website with images automatically with Opus and Dall-E
This script will build any website or collection of websites for you using Anthropic's Claude 3 Opus and OpenAI's Dall-E 3 which you can host online and share with others.
Code files for this project at Patreon: https://www.patreon.com/posts/opus-and-dall-e-102506058
Chatting with Opus about mysterious events in history. Modified to use OpenAI Whisper API.
Download code files for this project at Patreon: https://www.patreon.com/posts/voice-chat-with-102438582
I wanted to try something different for fun with this video, You can find the Code review at the end. Please let me know how you like this format
Older GPT-4 vs new GPT-4-TURBO vs Opus and Gemini as well as random models from LLM arena
We will test all 4 models with 4 difficult math problems plus a difficult coding problem with function calling
You can become a Patron and download all my project files: https://www.patreon.com/echohive42
Claude 3 Opus function calls but GPT-4 writes the tools' definitions automatically
Claude 3 Opus function calls but GPT-4 writes the tools' definitions automatically
Code files for this project at Patreon: https://www.patreon.com/posts/claude-3-opus-4-101811870
Opus & GPT4 Ultimate web research agent with Perplexity and Exa search and entity extraction
This agent can use either OpenAI GPT-4 or Claude 3 Opus along with Perplexity and Exa APIs to perform tireless research for you as it entity extracts important inbformation and combines your entire search into a single json file.
Opus Auto Coder with custom COT prompts and multi file generation
In this Auto Coder we can use simple commands to generate multiple alternative files in parallel and select from multiple custom prompts for each turn of generation. This Auto Coder works with Anthropic's Claude 3 Opus model
Voice chat with Claude 3 Opus using local faster whisper with ability for user to interrupt anytime
We are using the faster whisper local model for this project. to find out more about how the code works please watch this video: https://youtu.be/PuDwe9lpJWg
Auto code writing, executing, debugging Opus Auto Coder with Docker and Local mode
This Auto Coder can automatically run code within a Docker container or using subprocess. it automatically manages Docker containers. Do make sure to pay attention to your hard drive space and that the containers and images are being cleared properly.
Build apps with Claude 3 Opus and GPT 4 at the same time
This is the newest version of my Auto Coders which allows you to build projects with both models simultenously. You can also choose to work with a single model as well.
Claude 3 Function calling with streaming responses explained
Claude 3 Function calling with streaming responses explained
Code files for this project(all files) at Patreon: https://www.patreon.com/posts/claude-3-calling-99839998
showing how to make calls to Claude API regularly and as streaming responses. This video also has claude unified class which simplifies everything
Code files for this project(all files) at Patreon: https://www.patreon.com/posts/claude-3-unified-99755565
Can we improve GPT's logical reasoning with system messages?
In this video we will try to improve GPT's logical reasoning by trying out different system messages.
NOTE: Using gpt-4 with combinations file can cost 2-4 dollars per run!
Why abstractions are useful | Openai unified API abstracted class
In this video, the concept of abstraction in coding is explained as a method of grouping a sequence of events under a single name to simplify and reduce repetition, using the example of creating a class to handle OpenAI API calls efficiently. It also discusses the benefits and potential complexities of abstraction, emphasizing its utility in making code more reusable and manageable.
This video showcases how to build an archive researcher tool using GPT for querying and managing research papers, featuring real-time search, TFID vectorization, and database management techniques
GPT Meets Pygame: AI-Driven Python Game Experience
In this video, we take a look into a GPT-powered Pygame project featuring dynamic interaction between two circles. The focus is on technical solutions for threading and GPT API calls to ensure the game runs smoothly without blocking operations. The project illustrates the use of parallel processing and AI-driven decision making in a gaming context, highlighting the practical application of advanced programming techniques such as real-time API communication and agent-based system design.
This app will retrieve the latest 10 LLM related papers and display them with a link. you can chose on or enter your own utlr and it will summarize and produce a short audio for the paper so you can stay on top of the latest advancements. Deployment step to railway also explained in the video
Function calling agents using Json Mode only and Unified OpenAI API
In this video, we will explore how to build function calling agents using Json Mode only. We will use openai unified API which I have created to make things simpler.
GPT Searches the web and generates research reports with entity extraction with Perplexity + Exa
Ultimate GPT Researcher uses GPT to query Perplexity and Exa to find answers and generate research reports in a conversational format. You need Perplexity and Exa API keys which you can set in OpenAI unified file but it will auto detect them from your env variables.
Entity extraction, coder, reviewer, code structure agent systems
We will use openai unified api object I have written to build 3 agent systems
Code files for this project at patreon: https://www.patreon.com/posts/code-files-for-97690265?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Openai API calls unified so you can build complex apps quickly|GPT4 + GPTVision + DallE + Embeddings
OpenAI unified brings together all the calls you need to build complex apps drastically reducing the line of code you need in an easily understandable way
Code files for this project at patreon: https://www.patreon.com/posts/code-files-for-97599872?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Auto Interview & evaluate in parallel using observer pattern, queue system, threading
Auto Interviewer is an autonomous interview agent which can perform interviews for any position. It has two processes which run in parallel to interview and evaluate and write to database in parallel asynchronously. It features threading for parallelization a queue system and observer pattern for automatic notification of the evaluator of every Q and A pairs. It writes to a sqlite database
Code files and Exclusive video for this project at patreon: https://www.patreon.com/posts/code-files-and-97449846
Local SQLite vectordb with multi modal image search using latest text embeddings large
We will build a SQLite vector database using images and their embedding representations so we can perform cosine similarity search and find the most relevant images based on user query
Testing Newest OpenAI models: Embeddings large, small and gpt-4-0125
In this video, we will be taking the new models text-embeddings-small, alrge and gpt-4 turbo preview(0125) for a test drive and perform some tests. we will also learn about how to get several hundred embedding calls simultaneously using asyncio and make simultenous calls to gpt using threadpool executor.
Code files for this project at patreon: https://www.patreon.com/posts/code-files-and-97282701
FastAPI + GPT Full stack webapp . uses websockets, deployed at railway
This is a full stack web app build using FastAPI and GPT api. It uses webscokets and regular fetch reguest along with only tailwind styling. I also show how to deploy it online. it is a super fun quick app you can use even for a minute and learn some concepts about python. You can try it out here: https://fastpython.up.railway.app/
Code files for this project at patreon: https://www.patreon.com/posts/code-files-for-97221294
chat or QA with auto generated personas, system messages are auto generated also
In Multi Persona Chat, we take in a user question and auto generate different personas who can answer that questions complete with full and rich system messages, you can then chat with a persona of your choice or change it to a different persona, or start over again
Code files for this project at patreon: https://www.patreon.com/posts/96954901
GPT + METAPHOR internet search without crawling or scraping
Metaphor offers a neural model powered search API which also returns the contents of the urls so you don't have to crawl or scrape the contents yourself. This video will be an introduction to Metapho API
Using Python, search and GPT to find meaningful text in Library of Babel
We will recreate the library of babel partially and search for meaningful sentences in it then use GPT 4 to return any meaningful text which we can find
Make the best of GPT-5 potential as a coder by preparing for it ahead of time
What GPT-5 will mean for coding in 2024. How to best prepare to make the most of it as a new or seasoned coder. Importance of AI assistance while coding using Github Copilot, Cursor IDE and chatgpt
Fun way to code with GPT | Writes/rewrites python files in real time inside your IDE
Live assist works by you typing comments in the python files you are working on, it auto detects changes then writes and rewrites the file according to your specifications
Curated answers with multi query web search with GPT 4 Turbo and Perplexity 70b online model
In this video we will get gpt and Perplexity API to work together for detailed and curated answers to user questions using the web search functionality of Perplexity API. Since perplexity already curates the search results we will have doubly curated final answers
Should you learn programming? | how to code with AI assistance
This is a quick video sharing my thoughts on the value of larning how to code and also offering some tips and tricks I learnt while using AI assistance when I am programming.
Automated Chat Evolution: Streamlit UI with Gemini, Mixtral 8x7b, GPT-4 Turbo & Perplexity in Action
In this video we will review how we can build streamnlit interface for unified chat both for auto chat between any two LLMs and for user chat with any of them
Any LLM auto chat & chat with any LLM. Gemini, Mixtral 8x7b, GPT-4,Gemini, perplexity 70b web search
In this video we will take a look at the Unified Chat python class which will allow us to built complex multi llm interactions. we will chat with any llm out of Mixtral 8x7b, Gemini, Gpt 4 turbo, perplexity 70b onine. we will also get any 2 of these models to auto chat based on prior system and initial message
Gemini Sales Agent with RAG. Gemini basics with chat and vision and GPT sales Agent
In this video we will cover quite a lot of topics. We will compare Gemini against GPT. Gemini basic with chat and vision. How to build a sales agent using both Gemini and GPT. We will also cover Retrieving sales agents using similarity search and vectordb.
In this video we will review an experimental memory idea I had, which uses dynamic and parallel summarization along with secondary memory to keep the main memory for the GPT 4 chatbot much smaller than regular memory
Assistants API memory management, Unstructured API, Instructor and more
In this Live stream we will take a look at memory management for Assistants API, Unstructured API for automated data extrAction from all types of files, review instructor and why it is so good and other cool stuff time allowing
In this video we will overcome some of the limitations of threads in Assistants API and implement multiple memory management strategies including trimming the thread messages, summarizing and keeping messages and more advanced full summarization option at the end.
Email categorization and Fantasy game using instructor and GPT
In this video we will use Instructor library to create a python project which categorizes work related emails and another project where we create fantasy characters and let them duel until there is a single winner.
Multi Modal language tutor using GPT-4-TURBO, Whisper, OpenAI Text to Speech
In this video we will explore a multi modal language tutor which helps you translate any request between any number of languages automatically from Voice or text input. Each response is saved to a nicely structured markdown file and each audio response is saved to an mp3 file for future review.
GPT-4 Turbo retrieval documentation maker converts documentation into easily chunkable JSON format
Docs maker will take in documentation and will use the new GPT-4 turbo to convert it into easily chunkable and more accurately retrievable JSON object. We will explore 3 files from basic to more advanced. in the second and third files, we will try to iteratively improve the generated documentation.
JSON Mode experiments, writing books and gettings tons of structured data out of documents
In this video we will explore a project in which we will write a Python tutoring book with explanations, code examples and more. In another file we will stress test Json Mode by extracting many strings, lists and dictionaries worth of structured data from a document
Recommending the right product to user. only JSON mode. No Function calling. with and without RAG
GPT Sales Agent talks to the user in a friendly and kind manner in a regular chat and understands user preferences and only at the right time recommends to the user the product they are looking for. We will use only JSON mode and no Function Calling which simplifies things quite a bit. We will explore 2 versions one without RAG and one with Vector search and RAG. We will also use multiple search terms returned by the GPT to search over our database to find the right product for the user.
echohive GPT build tips and tricks, working with csv and json files for retrieval
In this video we will take a look at how I built the echohive video finder GPT along with some of the challenges I faced with retrieval and how I solved them dealing with CSV and JSON files.
Auto Scientist extracts the essence from a list of URLs and generates Original and Novel ideas
GPT-4 Auto Scientist can extract the essential information from a list of any URLs and Arxiv papers then iteratively generates original ideas based on this information.
Looping GPT 4 Vision and Dall-E 3 for iterative image improvements
In this video we will take a look at how we can get a description for an image from GPT 4 Vision and get Dall-E 3 to generate and image based on that description. then we will create a loop in which GPT 4 Vision tries to improve the prompt based on user instructions while staying true to the original image. We also review the OpenAI Dall-E 3 Cookbook
Assistants API explained completely A to Z step by step with code examples
In this video we will dissect OpenAI's new GPT 4 Turbo Assistants API entirely in 4 files from basic to most advanced. in the 4th file we will create an assistant with 2 custom functions, code interpreter and retrieval.
In this video we will narrate a video using gpt-4 and turn it into speech, we also explore 2 other examples on how to take snapshots of websites and get gpt-4 to describe them or offer improvements. we will also get snapshots of two websites and build sci fi story with them.
Everything you need to know with code examples. All new features covered
In this video we will explore how we can use the new OpenAI library to make calls to gpt-4-turbo with and without streaming, new function calling, image generation api with dall-e-3, GPT 4 Vision API, text tp speech API, and Assistants API.
Auto Self Fine tuning GPT 3.5 turbo iteratively improves itself
This script will allow us to let base gpt 3.5 turbo model to generate a fine tuning dataset which we then use to fine tune it, then use the new models to iteratively generate new datasets and keep going until a final fine tune is reached.
Instructor library automates function definition generation and parsing of the data from GPT function calls and gives you clean and easy to use data rather than dealing with complex json objects and parsing issues.
Story Maker gpt with structured output and Everything GPT Course intro
In this video we will take a look at a method which produces parsable structured output without using Function Calling. Which worls well even with gpt 3.5 turbo. I also talk about the new course I am working on. I would greatly appreciate your feedback
Memwalker advanced tree retrieval without VectorDB Python code inspired by the paper
in this video we will be looking at my personal interpretation and custom code implementation of Memwalker retrieval system in Python using gpot 3.5 turbo for summarization and GPT-4 for reasoning and, navigation and retrieval. Memwalker is a tree like retrieval system which doesn't rely on embeddings or VectorDBs. I mentioned in the video that the "visited keyts" might be off but I believe it is working correctly -tho I am not 100% sure :)
Run any Open source LLM on your computer. Mistral or Llama or others. Python inference code included
In this video we will use Lmstudio to download and run Mistral 7b on our computer without the need for a graphics card. You can also download and run any open source model.
Build GPT apps visually with Rivet drag and drop UI. Python code also included
In this video we will explore visual programming desktop app Rivet. Rivet allows you to use nodes and connect them to build gpt-4 powered apps. You can customize your API calls as you would in your python script. It seems very flexible and powerful. We will build an application which builds tower defense games. I will also review the python code which does the same thing. We will be using both gpt 3.5 turbo and GPT-4.
Automated fine tuning gpt 3.5 turbo for Function calling. dataset generation is also automated.
In this video we will use Llama Index to automate both dataset generation and fine tuning of gpt 3.5 turbo to use better function calling capabilities. this will allow gpt to output better formated structured data. This code allows you to generate as many fine tuning examples as you like.
Multi Agent AutoGen and Group Chat implementations step by step walkthrough
We will take a look at the Multi Agent and group chat interactions with AutoGen library by Microsoft. how we can use it to easily create agent interactions with OpenAI's GPT-4. We will also learn about how to set llm config, and set up Docker for safe execution of code blocks and function calling.
AutoGen Introduction in 5 files step by step including how to enable Docker
We will take a look at the AutoGen library by Microsoft in this video on how we can use it to easily create agent interactions with OpenAI's GPT-4. We will also learn about how to set llm config, save conversation history to a file and set up Docker for safe execution of code blocks.
GPT-4 auto prompt generation for GPT 3.5 turbo 16k experiment with evaluation
In this video we will explore an experiment I made using GPT-4 to generate and iteratively try to improve on prompts for gpt 3.5 turbo 16k for a specific coding task. Results weren't very promising but the code has many interesting ideas in it.
Voice controlled Auto AGI with swarm and multi self launch capabilities
This is a voice control update to Auto AGI. Now you can control it with your voice. transcription is handled by OpenAI's whisper API. Auto AGI uses GPT-4 function calling and can launch multiple instances of itself and also has a special swarm ability. links to previous videos are in the description for code review and other functionality
Voice control multiple GPT agents at the same time
In this video we will review a project I built which lets you control multiple single action Function calling GPT-4 agents simultaneously. You have to run multiple instances of the agents yourself. You can do this by splitting your terminal.
Fine tuning GPT 3.5 turbo to do highly complex tasks with minimal system message
in this video we will learn about many things, how to download articles from wikipedia, how to use gpt 3.5 turbo instruct, how to get GPT-4 to evaluate sumamries of documents, how to create a fine tuning dataset both for training and validation. how to fine tune a model, how to cancel fine tunes, how to view updates on the fine tuning process and how to delete fine tuned models. We will also look at the new fine tune UI at OpenAI playground.
New GPT 3.5 Turbo Instruct model released. Quick start guide
This video easy and simple quick start guide on how to start using the new OpenAI Gpt 3.5 Turbo instruct model which performs amazingly well! This is a completion endpoint so no complex assistant, user messages necessary, just prompt, completion pairs!
3 gpt-4 agents communicate to build application code. Includes Architect, Coder, Reviewer
Multi Agent Coder is an experiment. It sometimes works great and sometimes not so much. It has 3 agents which are initialized form a GPT-4 API class which communicate with one another in turn to write code and update it by review and redesign.
GPT-4 Auto AGI Agent gets communicating swarm intelligence as a tool upgrade
We will look at a new upgrade to Swarm Auto AGI which gets communicating swarm intelligence as a tool along with list file, create directory. It also has new commands to switch between gpt 3.5 turbo 16k and GPT-4. You can now add a new system message with a command and remove the last system message as well.
Swarm Auto AGI is a autonomous agent framework which can launch multiple instances of itself with open interpreter if needed and solve separate tasks or work on the same problem. Open interpreter is an open source code interpreter. Always be careful when auto executing code with a code interpreter!
GPT-4 Codes near perfect Llama Index code using Vector DB from documentation
We will learn about how to crawl a documentation website. Extract self sufficient code blocks along with generated explanations and create a dataset of the entire Llama-Index documentation. Then we will embed it and search over it using similarity search and let GPT-4 generate entire Llama index applications from simple descriptions
Communicating Swarm Intelligence prototype with GPT
A prototype of a GPT based swarm intelligence system with interagent communication for building future swarm systems. I will be building more complex swarm systems in the future.
Prototype of a GPT based swarm intelligence system as example to inspire more advanced swarm systems
A simple prototype of a GPT based swarm intelligence system for building future swarm systems. This is more like an Ensemble approach which keeps a lot of the complex aspects simple for understanding and insights purposes. I will be building more complex swarm systems in the future.
Auto generate fine tuning data for GPT 3.5 or other open source models from a single prompt
We will look at a system which automatically generates fine tuning data for GPT 3.5 Turbo or other Open source large language models such as Llama 2 from a single, simple prompt
How to install the requirements for Talking with multiple PDFs App
In this livestream, we will look at the requirements and how we can install them to get this app running.
This is for video: https://youtu.be/qiRSx7RnGjA
GPT-4 Codes near perfect Langchain code using Vector db from documentation
We will learn about how to crawl a documentation website. How to detect if code exists in a given file using gpt-3.5-turbo-16k, extract self sufficient code blocks along with generated explanations and create a dataset of the entire Langchain documentation. Then we will embed it and search over it using similarity search and let GPT-4 generate entire langchain applications from simple descriptions
We will build a ChatGPT like chatbot which uses langchain, openai GPT-3 and OpenAI embeddings and Wikipedia API to search over an uploaded document and answer questions using relevant context. Includes a Streamlit user interface with a file uploader.
GPT agent that actually works and writes its own tools
Today we will test the Auto AGI agent I have built which actually works and writes its own tools. This implementation uses pure OpenAI function calling
Local Vector DB, from user input, from urls, from files
We will build a local vector db using OpenAI ADA-002 embeddings from user input, from a list of urls and from files in a folder. We will use langchain for text splitting and embedding
Contents of this podcast is entirely AI generated. So bring a bit of skepticism along with your curiosity.
In this episode of echohive podcast we talk about AI takeoff, evolution, singularity and possible positive effects of AGI. We also consider what happens if no takeoff happens.
We will explore an idea to be able to defend against dangerous and adversarial prompts which tries to get the model to generate objectionable and undesirable content.
We will use OpenAI's Whisper and GPT-4 ro summarize and extract key points along with actions items and sentiment analysIs for meeting audio or any other audio files.
History of AI and Python. AI generated podcast experiment
In this experiment we will listen to two GPT-4 discussing the history of AI along with Python programming language and large language models. Please let me know what you think of this as in the future I will attempt to get the system to discuss the topics more in depth and longer
In this video we will review an app which auto downloads 10 arxiv papers based on a search query, lets gpt to decide on the most promising one then gets gpt to generate a new promising search query to continue the research.
In this experiment we will listen to two GPT-4 discussing the intersection of Continuum hypothesis about mathematical infinities and AI. Please let me know what you think of this as in the future I will attempt to get the system to discuss the topics more in depth and longer
We will learn about how we can build a chatgpt full stack webapp with message histories using FastAPI, MongoDB, htmx, Tailwind CSS needing only very little JavaScript. We will learn about how we can create a Mongo databse using railway, connect out web app to it and also how to deploy our webapp at Railway. Enjoy!
NOTE: Uses a lot of Tokens!!!
In this video we will explore tournament style code elimination with Elo rankings combined with a Genetic algorithm approach to see if we can produce better code.
GPT-4 Scientist generates original ideas from research
we will explore the new app that I built which can read arXiv papers, Wikipedia articles or custom provided knowledge and generates novel and original ideas based on essential information extracted from knowledge base.
How to build and deploy a GPT chat web app. with conversation history deployed on railway
We will continue to learn about how we can build full stack gpt powered web apps using OpenAI api and deploy them on railway apps. We will be using gpt 3.5 turbo, python FastAPI, html, css and JavaScript.
In this video we test gpt's comprehension ability over different context lengths while inserting certain information at different positions in the context.
Full stack GPT Webapp summaries of latest Arxiv LLM papers. Quick deploy on Railway
In this video we will see how we can build a full stack webapp powered by gpt 3.5 turbo 16k and deploy it on railway. This project includes all the files you can deploy online right away. We retrieve latest papers related to Large Language Models using arxiv api. then summarize the contents to a csv file and use the file to generate the contents for the webapp which is built with html using tailwind css and pure JavaScript backed by A FastAPi Python backend. we also save the OpenAI embeddings of the documents to the csv and there isa search.py file for doing similarity search over the paper summaries.
GPT Prompt Engineer Agent creates best prompt for a task with prompt tournament
We will take a look at a Github Repo which uses GPT 3.5 turbo and GPT-4 with Elo ratings and a tournament to generate the best prompt for a given task. This is an Agent which automates prompt engineering by using a Generator GPT api call and a GPT which ranks candidate prompts
Streaming responses with GPT Function Calling. Structured data with GPT 4 and Gpt 3.5 turbo
In this video we will focus on how we can get streaming responses using OpenAI function calling. We will also look at how we can use functons calling only to extract structured data out of gpt 3.5 turbo and gpt4. We extract summary, important points and key word out of a document
Streamlit's new chat elements with streaming responses and tiktoken conversation history limit
We will take a look at the new Streamlit chat elements to build conversational apps. We will also implement streaming responses using OpenAI api with gpt 3.5 turbo. You can also use gpt-4. We will implememnt a conversation history for the chatbot and use tiktoken to limit the max tokens so we dont go over the gpt's context limit. other topics include request timeout and using Tenacity for retrying functions which returns an error.
OpenAI Function Calling stress test with 20 functions and 100 questions
we will test chatgpt function calling first with 10 simple functions and then with 20 more complex functions. each time concurrently asking it 100 questions to see how many times gpt 3.5 turbo can choose the correct function.
Chatgpt Auto Builder codes multi file applications from basic descriptions
Chatgpt Auto builder & coder uses either gpt-3.5-turbo-16k or GPT-4 to call functions to build complex multi file apps from simple descriptions. You can work on your projects iteratively and resume your projects as well. It uses pure python and OpenAI API to make two consecutive calls to functiona calling variants of GPT to code entire applications.
Function calling Auto Coder App Maker builds multi file apps, resumes projects
Function calling Auto Coder App Maker can manipulate the file system using functions. It can read and write files, create folders, delete files and ask user questions when necessary.
We will explore Function calling for a function which returns html, css and JavaScript files within a single call to GPT-3.5-TURBO or GPT-4. We will also be coding to see what else we can accomplish
Langchain agents in JavaScript. All built in Agent types
In this #langchaintutorial video we will explore all built in Agent types with #langchain #js :
MRKL Agent, Conversational Agent, Plan and Execute Agent, Structured Tool Agent.
Function calling with Chatgpt (gpt-3.5.turbo, GPT-4) in depth exploration
in this video we will explore the new gpt-3.5-turbo-0613 and GPT-4-0613 models' function calling capabilities. we will look at a basic example at first and then a sql query generating code example and finally a concurrent summary generating arxiv scientific paper explorer
Chatgpt Web App boilerplate with streaming responses built with html, Tailwind css and JavaScript
In this video we will learn about how we can build a chatgpt Web App user interface boilerplate which can be modified for future designs built only using html, tailwind css and JavaScript which displays streaming responses. We will also take a look live server and live preview to be able to quickly view our web app while building it
NOTE: I meant to say "live preview" extension when I said "live share"
Auto Coder codes a GPT text adventure game, web automation tool and music synthesizer
In this video we will use GPT Auto Coder to code a text adventure game which uses gpt-3.5-turbo, a web automation tool which automatically searches google, saves a snapshot of the search results and extracts the text and saves it to a file, a very simple music synthesizer. ALWAYS be CAREFUL when auto executing AI generated code!!
Langchain & JavaScript deeper look at chat models, streaming responses, embeddings and more
in this video we will be taking a deeper look at how we can use chat models like gpt-3.5-turbo and GPT-4 with JS Langchain along with streaming responses, embeddings, rate limiting and more.
Langchain in JavaScript, step by step Introduction
in this video we will learn about how we can implement Langchain in JS step by step with 6 files. We will talk about making calls to gpt-3.5-turbo or GPT-4, Prompt templates, memory, Langchain Agents, Streaming completions and more.
Custom LLM Agent with tools from scratch with python and vanilla OpenAI
in this video we will look at how we can build a custom agent which uses tools such as google search built only with Vanilla OpenI and pyton. We will also learn about tiktoken for counting token usage, termcolor for colorful printing in the terminal, py expression eval for calculator.
SCIKIT-LLM: Scikit-learn meets Large Language Models
This video is a quick look at this cool repository called SCIKIT-LLM which allows you to do classification, vectorization, summarization and more:
https://github.com/iryna-kondr/scikit-llm
20 documents summarized simultaneously with gpt 3.5 turbo and asyncio
In this video, we will summarize 20 Wikipedia articles about artificial intelligence topics which we get using requests library in python. then we will break them into chunks of 500 tokens. Use Asyncio to make parallel calls to gpt-3.5-turbo while avoiding rate limit error with a delay tactic. we will also learn about termcolor to print colorfully in the VS Code terminal and about backoff library which also implements exponential backoff retry.
100 Gpt-4 API calls simultaneously with asyncio with Tenacity retry
in this video we will make 100 simultaneous (divided into batches of 50) to OpenAI's GPT 4 API endpoint using python's asyncio library with Tenacity library to retry in case of rate limit errors. We will be generating 100 python function. we will also be talking about termcolor for colorful prints in terminal and pretty js/json VS Code extension to make json files more readable.
2 GPT agents work together to solve programming tasks for you with streaming responses
in this video we will be looking at a Vanilla OpenAI implementation of 2 gpt-3.5-turbo or GPT-4 agents working together to solve a programming task you have presented with streaming responses. You can modify the system messages to make this work for whatever use case you are imagining.
ChatGPT web UI locally for GPT-4 using FastAPI and JavaScript
in this video we will take a look at a Chatgpt web user interface replica I built for GPT-4. This app uses only GPT-4 unless you modify the code. It is very fast and built with Python,m FastAPI, HTML, CSS and JavaScript.
GPT-4 Auto Coder 3 with Code Interpreter. Runs locally with all Python packages
In this video we will be looking at the version 3 of the GPT 4 Auto Coder apps that I have been building. This one also has a Code Interpreter and is able to auto run the code it generates and is able to iterate on it by using th eerror messages, regular terminal output and also user feedback. Always use caution when running AI generated code automatically!
Langchain D&D game simulation with Dall-E 2 for story images
in this video we will use langchain library's example for 2 player D&D agent simulations. We will also use Dalle-E 2 to automatically create images for the ongoing story and we will use matplotlib to display the images as well.
Synthesize answers over multiple individual Vector Indices with Llama index composable graph
In this video we will explain how to use the LAMA Index for building composable graphs. These graphs allow you to build indexes on top of other indexes, which is beneficial when you want to query multiple indexes at the same time, something that would be difficult with traditional single index similarity searches due to the limit of chunks that can be fed into the large language model. We will also learn how we can use chat models such as gpt 3.5 turbo and gpt-4 with llama index
In this video, we will delve into the semantic search functionalities of Llama Index (GPT index). The video covers four specific aspects: index storage, index naming, login dynamics with the use of info and debug for additional insights, and token usage. we will also explores the innovative concept of using a support vector machine (svm) for retrieving similarity results.
Llama Index ( GPT Index) step by step introduction
In this video we will learn to use llama index, formerly known as gpt-index to store documents to vector store indexes and then ask questions about them. We will also learn how to add multiple documents to the same index and create multiple indexes both using from documents functionality and nodes. We will also learn how we can customize the llm and the use the prompt helper for setting token limts etc...
QA over documents with Auto vector index selection with Langchain router chains.
In this video we will learn how to use langchain router chains to be able to automatically select the right index which matches our questions. We will take a look at the code for this both for streamlit and terminal versions. We will also learn about auto selecting prompt chain as well which selects the appropriate prompt for the user question.
Build WebApps from a csv file and deploy to railwayapp. How I built the echohive app
In this video we are going to learn about how we can use a simple csv file to dynamically generate an entire webapp with search functionality. I created www.echohive.live for my viewers to easily be able to search all my videos and find code download links conveniently. We will talk about how to extract youtube video information using Google Youtube API V3, turn them into csv file and programmatically generate html and deploy it to railwayapps.
GPT-4 Auto Coder 2 reads docs from URL links and Auto Debugs
This version of Auto Coder uses PlayWright from langchain to read contents from documentation using URL links. Uses GPT-4 API call to extract the essential information and then uses Pyton Repl agent from langchain to take in user instructions to auto code, auto run code and auto debug code repeatedly until hopefully it generates working code for user requirements
GPT-4 Agents cooperate to solve tasks. OmniSolver webapp
This app is based on CAMEL framework and Langchain implementation to use two GPT models(agents) to cooperate to (try to) solve any given task. this code uses OpenAI chatcompletions endpoint. Both gpt-3.5-turbo and GPT-4 models can be used. Two agents simulate a conversation while solving given tasks
GPT Simulated Environments: gpt-3.5-turbo plays blackjack in OpenAI RL Gym environment via langchain
In this video, we will be building upon a Langchain implementation of a simulated reinforcement learning environment in which chat GPT models can play blackjack in. Code will feature Gymnasium library which is formerly knows as OpenAI Gym for reinforcement learning agents.
Code download is available for Patreon supporters here along with 50+ other project files:
How to install the requirements for Talking with multiple PDFs App
In this livestream, we will look at the requirements and how we can install them to get this app running.
This is for video: https://youtu.be/qiRSx7RnGjA
Meta Prompt: Self Improving GPT-4 Agent
in this video we will be implementing the META PROMPT agent from this article: https://noahgoodman.substack.com/p/meta-prompt-a-simple-self-improving
We will be looking at two different implementations. One which uses a single loop and another which uses the two loops as in Meta Prompt. We will be using Vanilla OpenAI API onlyusing GPT-4 but you can use gpt-3.5-turbo as well.
Effortless Note Organization with GPT-4 & word count constraints
In this video we will explore how we can use GPT-4 to organize our notes about many different topics into categories and write them to files with the associated category so we can have them all nicely separated and structured. We will also talk about how we can attempt to get GPT-4 to respect user given word count limits when generating content or when expanding on ideas.
This video is a tutorial course on prompt engineering in Python.
we will explore 16 different prompts and how to engineer them for use with OpenAI chatcompletions API with gpt-3.5-turbo and GPT-4
In this video we will be reviewing the code for how we can read files and convert them to speech or generate poetry with GPT-4 using OpenAI API and AWS Polly service.
Generative Agents: Interactive Human Behavior Simulation with Langchain and OpenAI GPT
Step into the world of interactive human behavior simulation with this detailed tutorial on Generative Agents. Dive into Python coding, harness the power of LangChain library Faiss for vector storage, and explore the capabilities of OpenAI's GPT-3.5 Turbo or GPT-4. Transform your projects with this cutting-edge technology and create immersive experiences like never before!
Build your first AI powered App Crash Course part 2
Purpose of this series is to get you up to speed with some basics so you can start building your own Ai powered apps.
This is a series which focuses on learning how to build AI powered apps as quickly as possible. In this video we will learn about error handling and debugging and about comments, docstrings. We will also build our first App using OpenAI API.
Purpose of this series is to get you up to speed with some basics so you can start building your own Ai powered apps.
This is a series which focuses on learning how to build AI powered apps as quickly as possible. In this video we will learn about the basics of dealing with VS Code, terminal commands, PIP pachages and virtual environments using Conda.
GPT 4 auto generates Stable Diffusion prompts and auto improves them from user feedback
we will get OpenAI's GPT-4 to generate text to image prompts from user ideas, then generate stable diffusion images and improve the prompts with GPT-4 using user feedback mechanism. we will use openai api, gpt4 and stable diffusion model v1.5
Chat with and Summarize PDF documents with Langchain and OpenAI
In this video, we will look into how we can build a system which allows us both to summarize and chat with PDF documents using lanchain library and OpenAI API gpt-3.5-turbo.
We will be chatting with Nvidia annual report, GPT-4 technical report and the "attention is all you need" paper which gave rise to transformer based large language models with attention.
GPT-4 teaches programming. learn Python, JavaScript, Go, Rust, HTML, CSS with an AI tutor
In this video we will do a demo and code review of the streamlit app I built which teaches programming in Python, JavaScript, Go, Rust, HTML, CSS. This webapp uses gpt-3.5-turbo or gpt4 to make OpenAI API calls while the UI is taken care of by streamlit. I hope you will find this app useful
In this video w will review a different version of the GPT 4 Auto Coder Self Improver which executes the provided or generated code automatically and then feeds the error back to OpenAI gpt API along with the code to get improvements and bug fixes in as many iterations as you like.
I have put in a safety mechanism which prompts you to review the code before executing as self executing code can be harmful to your computer.
GPT-4 Multi threaded Auto Coder for simultaneous multiple code generation and improvements
We will review the code for Multi Threaded GPT-4 Auto Coder which allows you to start as many coding and improvement processes in parallel simultaneously. You can code a calculator, a pygame and a financial computations class all at the same time and improve them for as many iterations. Auto Coder uses OpenAI API and makes calls to either gpt-3.5-turbo or GPT 4 model. User can specify which model they need to use in the code.
GPT-4 Auto Coder #2 Code Review and improved Streamlit WebAapp UI
we will deep dive into the code for GPT 4 single threaded Auto Coder built by using vanilla OpenAI APi and improved streamlet webapp UI.
You can now type or paste your already existing code in the browser with the Streamlit user interface and the improved iterations of the code generated by GPT4 will display in the browser as copiable markdown.
We learn about how we can get GPT-4 or GPT-3.5-TURBO to write code for us and iteratively and automatically improve on it. We will be using vanilla OpenAI API. GPT 4 will write code for us from a simple idea or it can generate the idea itself too and write the code for it. Alternatively you can specify code files and let GPT4 improve them automatically. You can run as many of these processes as you like in parallel because we make concurrent calls to OpenAI API
GPT-4 Auto Codes websites with HTML, CSS and JavaScript
In this video we will explore an interesting idea on how we can make consecutive calls to GPT 4 firt to generate an idea for a websire design then with that idea within our next prompt to generate the HTML and CSS codes and automatically write them in approproate files and folders. Then we will auto launch the website in our browser and ask the user of they like it. If the user answers "no" then we will call GPT 4 again for new html and css files. When the user like the design then we will call GPT-4 again for the JS code and ask the user again for feedback. This is very cool in my mind to use OpenAI's API to create automated coding processes. I hope this will inspire you to build your own ideas.
Code files are available for download for Patreon supporters along with 35+ other project files:
GPT-4 API with streaming responses, chat history, streamlit, gradio UI, JavaScript, FastAPI webapp
This video will explain how to use GPT-4 api, create an evolving chat history which respects token context limits by droppping earlier conversation when necessary. We will also learn about how to stream responses to the terminal and with the Steramlit UI. We will build two different gradio UIs and also a webapp using HTML, CSS, JavaScript, Python and FastAPI.
Code files are available for download at Patreon along with 35+ other project files: https://www.patreon.com/posts/gpt-4-api-with-80312421?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
GPT-4 codes ChatGPT chatbot with memory webapp with FastAPI and JavaScript
In this video we will be overviewing the code entirely written by GPT 4 for 3 different UI implementations for a chatgpt (gpt 3.5 turbo) API from OpenAI.
The bot will have conversational memory and will respect token limits by dropping earlier conversations when appropriate. So you never have to worry about errors related to Token limits.
GPT-4 codes animations with JavaScript and Python Pygame
We will get GPT 4 from OpenAI, using ChatGPT plus model to create HTML canvas animations using JavaScript. We will create 4 different files with increasing complexity ( We will use Liver Server extension in VS Code for serving the local webpage) and eventually, we will ask GPT-4 to convert it to Python code implementing the same animation using Pygame.
Bonus video is available for Patreon supporters: https://www.patreon.com/posts/bonus-content-4-80081583
Whisper API speech to text with OpenAI and Python speech recording
We will learn how to use OpenAI's Whisper API in this video to transcribe speech from audio files to text. We will also look at two different python code tutorials on how we can record speech into audio to use with Whisper API
Code files are available for Patreon supporters along with 20+ other project files: https://www.patreon.com/posts/whisper-api-to-80047022?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
In this video we will discuss the capabilities of gpt 4 in comparison to earlier models. We will also see how we can generate code using gpt-4 with ChatGPT plus access to generate html and css code for a working video website.
Code is available to Patreon suporters for 20+ projects and blog posts: :https://www.patreon.com/echohive42
ChatGPT with any YouTube video using langchain and chromadb
this code uses gpt 3.5 turbo api (chatgpt api), langchain, chromadb to be able to ask questions to a youtube video's transcript. You can chat with any YouTube video you wish!
Code files are available for download along with 20+ other projects at Patreon: https://www.patreon.com/posts/chatgpt-with-any-79801283?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Langchain Async explained. Make multiple OpenAI chatgpt API calls at the same time.
Learn about how you can use async support in langchain to make multiple parallel OpenAI gpt 3 or gpt-3.5-turbo(chat gpt) API calls at the same time and have significant speed up in your applications.This video talks about Async support for llm calls, chain calls and agent calls with tools such as google search.
Code files are available for download at Patreon: https://www.patreon.com/posts/langchain-async-79611987?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
2 CHATGPT APIs talk to discuss ideas with a gpt 3.5 summary. Gpt 3.5 turbo conversations and summary
we get two Chatgpt OpeanAI APIs (gpt-3.5-turbo) to talk to one another to discuss ideas while at the end we summarize it with text davinci 003 OpenAI API
Code files are available for download for Patreon supporters: https://www.patreon.com/posts/2-chatgpt-apis-3-79547105?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
GPT 3.5 Turbo OpenAI API with auto earlier message removal using Tiktoken
GPT 3.5 Turbo OpenAI API with auto earlier message removal using Tiktoken
Code file is available for download for Patreon supporters: https://www.patreon.com/posts/gpt-3-5-turbo-79475116?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
AI UPDATE: Prompt engineering, Text to design, AI chatbots with GPT 3, Langchain update and more.
AI UPDATE: Prompt engineering, Text to design, AI chatbots with GPT 3, Langchain update and more.
echohive Patreon: https://patreon.com/echohive42?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=creatorshare_creator&utm_content=join_link
Langchain: Auto evaluate llms and Agents. OpenAI Gpt 3 evaluation using Langchain and custom prompts
Learn how to evaluate performance of llms such as OpenAI gpt-3 using langchain evaluation using custom prompts and datasets from huggingface.
Files are available for download at Patreon: https://www.patreon.com/posts/langchain-auto-79211882?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Langchain Summary and QA with Chromadb using OpenAI Embeddings and GPT 3 with token count
We will build 5 different Summary and QA Langchain apps using Chromadb as OpenAI embeddings vector store. We will use GPT 3 API to summarize documents and get answers to questions using context. We will learn how to count tokens and build a custom prompt for langchain. We will also learn to use different chain types such as "stuff", "map reduce", "refine" and "map rerank"
Files are available for download for Patreon supporters: https://www.patreon.com/posts/langchain-and-qa-79123754?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
How to install Unstructured for Langchain Document loaders. Langchain Unstructured tutorial
These installation steps for unstructured enables document loader to work with all regular files like txt, md, py and most importantly PDFs. You can also extract text out of images which is super cool. Also reads text our of any given list of URLs.
Instruction post is available for Patreon supporters: https://www.patreon.com/posts/how-to-install-79024617?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Langchain Chains and Chaining explained. Gpt 3 llm tutorial with streaming responses, multi input
We will learn about how to form chains in langchain using OpenAI GPT 3 API. How to build chains with multiple llm calls with multi input and multi output chains. We will also learn how to use streaming responses in langchain.
NOTE: YOU MUST HAVE "OPENAI_API_KEY" ENV VARIABLE SET FOR OPENAI GPT 3 API TO WORK
Custom langchain Agent & Tools with memory. Turn any Python function into langchain tool with Gpt 3
We will build a web app with Streamlit UI which features 4 Python functions as custom Langchain tools. Our agent will also have a short term conversational memory.
QA chat with a website, OpenAI embeddings tutorial: Use GPT 3 API and Openai ADA-2 Embeddings
We will learn about how to convert the text contents of a website in to OpenAI embeddings to be able to do similarity search for context for our question answering GPT 3 API.
File is available for Patreon supporters: https://www.patreon.com/posts/qa-chat-with-78647021?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Echo and Scotty try to figure out what ChatGPT is?
Langchain ChatGPT your documents challenge PART 2 with Gpt 3 and OpenAI embeddings and Streamlit UI
We will build a ChatGPT like chatbot which uses langchain, openai GPT-3 and OpenAI embeddings and Wikipedia API to search over an uploaded document and answer questions using relevant context. Includes a Streamlit user interface with a file uploader.
Files are available to Patreon supporters: https://www.patreon.com/posts/langchain-your-78469576?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Langchain ChatGPT your documents challenge with Gpt 3 and OpenAI embeddings and Streamlit UI
We will build a ChatGPT like chatbot which uses langchain, openai GPT-3 and OpenAI embeddings to search over an uploaded document and answer questions using relevant context. Includes a Streamlit user interface with a file uploader.
Streaming completions from GPT 3 API. Chat gpt like streaming text with Streamlit web app UI
We will learn about getting streaming responses from GPT-3 API instead of waiting for the entire completion. We will use Streamlit to build a quick user interface for this.
Code file is available for Patreon supporters: https://www.patreon.com/posts/streaming-from-3-78271909?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Python web application made easy and fast with Streamlit. Streamlit Python tutorial in VS Code
Discover the power of Streamlit with this Main Concepts tutorial! Learn how to create interactive web applications that can run at lightning speed using python and VS Code.
Patreon supporters can download the code file at: https://www.patreon.com/posts/streamlit-main-78235954?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Gpt index "Talk to Wikipedia" App with Streamlit, Gpt 3. OpenAI embeddings in Python, VS Code
"Experience OpenAI's GPT 3 and bring Wikipedia to life with cutting-edge "Talk to Wikipedia" webapp built with Streamlit, Gpt-index and Wikipedia API. Immerse yourself in a world of endless knowledge , enhanced by the latest OpenAI embeddings and gpt index. Includes a section examining VS Code debugging feature
Patreon supporters can download the code files here: https://www.patreon.com/posts/gpt-index-talk-78193169?utm_medium=sharesheet&utm_source=sharesheet&utm_campaign=postshare_creator&utm_content=join_link
OpenAI embeddings explained with code examples. how to use OpenAI Gpt3 ADA vectorized embeddings API
Learn how to utilize the advanced ADA vectorized embeddings API for more effective natural language processing and for advanced search and machine learning applications.
Langchain Gpt 3 Chat bot with memory. AI tutorial: 6 different memory types coded and explained
Explore the world of GPT-3 with a cutting-edge chat bot featuring advanced Chat Gpt like memory capabilities. With the Langchain tutorial, learn how to code and implement 6 different memory types for an enhanced user experience
Learn how to build an automated prompt engineering UI using GPT3 and Gradio in this tutorial. Utilize the power of GPT-3 to generate prompts for ChatGPT with ease and improve your workflow.
Join echohive Discord: https://discord.gg/PPxTP3Cs3G
GPT 3 searches the internet with langchain. how to use langchain tutorial: Agents & Tools explained
explaining lanchain agents, tools and the chain which allows gpt 3 to search the web or retrieve an answer from Wolfram Alpha or any other tool.
Join echohive Discord: https://discord.gg/PPxTP3Cs3G
Recipe app in 10 minutes with Huggingface transformers and Gpt 3 API. tutorial: learn by building
we will use python, huggingface tranformers API, gpt 3 API, Tkinter to build and app which detects food items from an image and then gets a recipe from gpt3. You can download the code from Patreon if you are a supporter: https://www.patreon.com/posts/recipe-app-file-77749194?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Gpt 3 tries verbal puzzles with self reflection prompt engineering
we will try to get Gpt 3 to solve some verbal riddles using prompting to induce self reflective analysis
Unleash the power of AI on your website with custom-built GPT-3 API integration
Seamlessly blend HTML, CSS, Javascript, and Python code to create a functional webapp that leverages the cutting-edge capabilities of the OpenAI API.
Code download for Patreon supporters: https://www.patreon.com/posts/chatgpt-ui-file-77559661?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
Chat Gpt UI built by Gpt 3 API responses directly in VS Code. Using HTML,CSS , JavaScript & FastAPI
We will be building a ChatGPT like user interface using OpenAI GPT api responses directly in VS Code. we will code HTML, CSS, JavaScript front end and Python backend with FastAPI.
Python tutorial: top 10 tips and tools explained in 5 minutes.
this video goes over Python list comprehension, enumerate method, itertools module, functools module, lambda functions, map function, filter function, python reduce function, python zip function, os.walk method
Creating Pytorch neural network and custom dataset: custom data + gpt 3 + github copilot
we will create a custom dataset and then use gpt 3 to code a Pytorch neural network. We will train the Pytorch NN on the data and print loss and accuracy. Finally we will test it on examples
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Kaggle with XGBoost, Gpt 3, chat gpt and Github Copilot:
We will be solving a Kaggle challenge using xgboost algorithm mostly written by gpt 3 in OpenAI playground. We will add additional code with the help of chat gpt and Github Copilot using VS Code as our IDE.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Lets learn Clojure programming language together: Clojure tutorial
I have no idea how Clojure works. We will go step by step and try to install and work with Clojure together :)
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
OpenAI Playground deep dive:How to use OpenAI Playground in depth
this gpt 3 tutorial video, we will cover all the options and buttons present in the OpenAI Playground interface with examples and in depth explanations.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Learn Javascript Objects with chat gpt. Javascript tutorial Part 3: Objects
In this video we talk about JavaScript Objects: creating objects, accessing object properties, adding and modifying object properties, iterating over objects
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
full video tutorial: https://youtu.be/5SqrahLVZtU
this video shows how you can get gpt 3 API responses and display them directly in your VS Code, taking advantage of syntax highlighting and more.
Gpt 3 API running directly in VS Code. implementting OpenAI API to display responses within VS Code
this gpt 3 python tutorial uses gpt 3 api and displays responses in Visual Studio Code directly. There is tkinter user interface in which we can enter our prompts and send it to gpt 3. we will build a realtime matplotlib plot to demonstrate a usecase.
Patreon supporters can download the code files here: https://www.patreon.com/posts/gpt-3-in-vs-code-78148884?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
How many unique names can gpt 3 generate without repeating itself
we will put GPt 3's ability to generate random unique names
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
it is easy!
AI assisted coding videos: https://youtube.com/playlist?list=PL611FKPtL864pFXwGbIcU4hvYHQBLd1kj
GPT 3 codes Python Turtle loops from user input with OpenAI API: Python gpt 3 Turtle library
Create any shape of turtle drawings with OpenAI GPT-3 and Python Turtle. In this video, we'll show you how to use OpenAI GPT-3 to generate code for Python Turtle. We'll then review the code gpt 3 returned and provide a simple graphical interface for executing it.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
a poem written by chat gpt about the significance of arrow functions in js
Fine tuning Gpt 3 with a custom dataset
we will use a webapp to generate dataset to fine tune gpt 3. we will cover everything about the gpt 3 fine tuning process.
patreon supporters can download the code files here: https://www.patreon.com/posts/fine-tuning-gpt-78149081?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
explaining javascript conditional statements tutorial: if, else and switch
this video is part of a many parts series teaching javascript entirely with shorts videos. welcome to the javascript tutorial with shorts :)
Creating a Gpt 3 Webapp with OpenAI API, FastAPI, html, JavaScript and Python
building a webapp API interface to gpt 3 with FastAPI and Python as backend, html and JavaScript as frontend. most of the html, js and python code will be written by openai playground.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
chat gpt poem dedicated to python list comprehension
I made chat gpt write a poem about Python list comprehension :)
I think it is cool.
Learn Javascript with chat gpt. Javascript tutorial Part 3: Arrays
In this video we talk about javascript arrays: Creating arrays, Accessing array elements, Iterating over arrays, Adding elements to arrays, Removing elements from arrays:
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Python vs JavaScript? chat gpt discusses it with itself. AI generated convo both text and audio
I personally love both languages!! :)
Javascript Assignment Operators explained
explaining javascript assignment operators.
this video is part of a many parts series teaching javascript entirely with shorts videos. welcome to the javascript tutorial with shorts :)
OpenAI playground will allow you to do almost anything you want with language. try it out!
link to full length video: https://youtu.be/KqwGdyBb-MM
How to use OpenAI Playground
we will be exploring how to use openai playground with the many examples openai has provided.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Javascript Logical Operators: AND, OR, NOT (&&, ||, !)
explaining javascript logical operators: && (and), || (or), ! (not
this video is part of a many parts series teaching javascript entirely with shorts videos. welcome to the javascript tutorial with shorts :)
Javascript comparison operators
explaining javascript comparison operators: equals to, not equals to, greater then, less than etc.
this video is part of a many parts series teaching javascript entirely with shorts videos. welcome to the javascript tutorial with shorts :)
JavaScript arithmetic operators
explaining javascript arithmetic operators: plus, subtract, multiply, divide and remainder
this video is part of a many parts series teaching javascript entirely with shorts videos. welcome to the javascript tutorial with shorts :)
javascript data types explained
explaining javascript data types: integers, floats, strings, booleans, arrays, objects.
this video is part of a many parts series teaching javascript entirely with shorts videos. welcome to the javascript tutorial with shorts :)
Learn Javascript with chat gpt. Javascript tutorial Part 2: functions
In this video we talk about javascript functions including declaring javascript functions, Function parameters, return values and Higher-order functions.
this javascript tutorial for beginners teaches you how you can use chatgpt to get clear and simple explanations to difficult concepts with code examples you can use and learn from
this video is part of a many parts series teachING javascript entirely with shorts videos. welcome to the javascript tutorial with shorts :)
javascript reduce method explained
javascript reduce method explained, which is super useful!! :)
check out my declaring variables tutorial: https://youtu.be/D122-szBLio
Chat gpt codes HTML document. learn html with chat gpt for beginners
Get started with HTML with this beginner's tutorial by Chat GPT: AI! In this video, Chat GPT: AI will guide you through the basics of HTML and help you create your own HTML file.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
javascript filter method explained, which is super useful!! :)
if you like this short check out my other videos @echohive
javascript map method explained
javascript map method explained, which is super useful!! :)
Learn Javascript with chatgpt. Javascript tutorial Part 1
Join chatgpt as it takes you on a comprehensive journey through the world of JavaScript. From the basics of variables and data types, to more advanced topics such as loops, conditional statements, and object-oriented programming.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
python reduce function explained by github copilot
@echohive we use AI tools to learn, code, build and have fun with python, javascript, chatgpt, gpt 3 and more
chatgpt teaches numpy: python numpy tutorial
In this video, you will learn all about the Numpy library, a powerful tool for working with numerical data in Python. We will start with the basics, such as creating and manipulating Numpy arrays, and then move on to more advanced topics such as performing mathematical operations, reshaping and transposing arrays, stacking and splitting arrays, and working with Numpy arrays and Python's built-in data structures. We will also cover advanced Numpy concepts and functions, such as broadcasting, masking, and linear algebra, and provide plenty of code examples to help you understand how these concepts work in practice
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
we showcase Chatgpt, an advanced artificial intelligence trained to solve programming problems on the popular Q&A site Stack Overflow. In this video, we demonstrate how Chatgpt can solve a variety of Python programming problems
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Link to full Tutorial: https://youtu.be/RyVTCljXO2E
Gpt 3 teaches Python Dictionary methods with Openai playground
We will take a look at all the built-in methods there are in Python for Dictionary manipulation. we will us gpt 3 to give us code examples. Hope you will enjoy! Please Like and Subscribe if you enjoy the content!
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Fun with html, css, javascript, fastapi using gpt chat and GPT 3
Join us for a wild ride through the exciting world of html, css, javascript, fastapi! We'll be using chat gpt and GPT-3 to create a webapp with FastAPI and Vanilla JS and learn how we can create many CSS style sheets from a single HTML file in seconds!
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Gpt 3 teaches Python List methods with openai playground
We will take a look at all the built-in methods there are in Python for list manipulation. we will us gpt 3 to give us code examples. Hope you will enjoy! Please Like and Subscribe if you enjoy the content!
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
gpt 3 Adventure Game Development with Python: Part 3 --Allowing User to enter a Custom Strategy--
- Adding input feature so the user will be able to enter a custom strategy to deal with the situation. - Fixing the readability of the response by using .split method.
- Managing the For loop for user entering a custom strategy
Gpt 3 teaches Python string methods with Openai playground
We will take a look at all the built-in methods there are in Python for String manipulation. Hope you will enjoy! Please Like and Subscribe if you enjoy the content!
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
gpt 3 python tutorial for beginners Part 3 object oriented programming
GPT-3 Teaches Python Part 3, Object oriented programming.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
chat gpt codes javascript animations using html canvas element
This video is just an attempt at creating Javascript animations using chat gpt only. Some of the stuff worked and some didn't.
I hope you will enjoy it.
Stable Diffusion: Generating Spectacular Visuals with GPT 3 Prompts
Explore the mesmerizing visuals created by GPT-3 when given the task to generate diffusion images. This video will take you on an exciting journey through these stunning images, as you discover new patterns and shapes. From abstract, to geometric, to peculiar, watch this video to get a glimpse of the beautiful possibilities available with GPT-3 and Stable Diffusion
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
gpt 3 Adventure Game Development with Python: Part 2 --Getting response in Python with OpenAI api--
We are getting the response to our prompt using OpenAI's api in this video.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Python sorting algorithms and complexity analysis with GPT-3
we will examine different sorting algorithms(bubble sort, selection sort, quick sort) and discuss why the time complexity matters. We will use gpt-3 and Github Copilot to program the sorting algos and put them to the test
Enjoy!
FastApi image gallery with Github Copilot and GPT-3
We are coding a image gallery entirely with Fastapi. Our helpers are Github Copilot and GPT-3 playground.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Here is another experiment coding with fastapi using gpt-3 assist.
We are creating three buttons which changes 3 different element's colors to a random color when clicked
coding a pygame simulation with AI assistance. No commentary just code and music. Python + vs code + github copilot + gpt-3 + chatgpt
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
90 percent coded by AI, this boid simulation features circles moving around randomly while they are trying to find a target. if they find the target, they grow in size and change in color depending on their size. pretty fun little project. Hope you will enjoy it.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Chatgpt and gpt 3 codes Tkinter canvas animations in marathon style
GPT-3 and ChatGpt takes turns coding tkinter canvas animations non stop in Python :)
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Chatgpt writes an entire functional custom game from scratch using python and pygame creatively limited by user feedback. This was incredible to see and I hope all of you will enjoy it too
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
I am using the OpenAI GPT-3 JavaScript sandbox to create the base code for the animated circles and then completing it in VS code using Github Copilot.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
gpt 3 Teaches Python for beginners Part 2 Functions/modules, Working with Files/Directories and OOP
GPT-3 Teaches Python P.2 3. Functions and modules 4. Working with Files and Directories5. Intro to OOP
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
gpt 3 teaches python for beginners Part 1. Basic Syntax and Data Types and Control Flow and Loops
GPT-3 teaches python. 1. Basic Syntax and Data Types 2. Control Flow and Loops
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Debugging the painting app with Gpt-3 codex and copilot only. Eraser debug was unsuccessful
GPT-3 edit was able to debug two of the problems but failed to debug the eraser problem. Second half of the video are the unsuccessful attempts at fixing the eraser.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Watch AI build a Tkinter painting app from scratch
Github copilot and OpenAI codex takes turns writing Tkinter painting app from scratch.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
AI codes TKinter GUI in Python using Github Copilot only
I only write the comments and the word import to get the Copilot started.
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
making circles bounce around with Python, Turtle library and Github Copilot...
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Solving StackOverflow problems with OpenAI GPT-3 and Github Copilot
lets see how good gpt-3 and copilot at giving us the right answers to out python programming problems :)
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
Kaprekar's constant and Collatz conjecture explained by AI
following links explain these interesting math mysteries:
If you find my content beneficial, consider supporting me on Patreon: https://patreon.com/echohive42
experimenting with Echoing with different generated images and descriptions. Check it out at www.EchoHive.live
Welcome to EchoHive updated :)
EchoHive uses Stable Diffusion along with a powerful AI language to help you with your prompts coupled with many other prompt tools to create the best images possible. check it out at www.echohive.live