CEO @HyperWriteAI, @OthersideAI - I make AIs do the impossible.
5 subscribed
Apr 10 • 4 tweets • 1 min read
Introducing `gemini-youtube-researcher` 📈
An open-source Gemini 1.5 Pro agent that LISTENS to videos and delivers topical reports.
Just provide a topic, and a chain of AIs with access to YouTube will analyze relevant videos and generate a comprehensive report for you.
This uses the new Gemini 1.5 Pro API that was released today.
It currently only supports listening to the audio content of videos. If anyone wants, please feel free to add support for video frames as well.
Apr 8 • 4 tweets • 1 min read
Open-sourcing `AI-Oracle`.
Generates better responses than Claude 3 Opus.
A very simple approach that combines the abilities of Claude 3, GPT-4, and Perplexity to provide better results than any could provide on their own.
Seriously -- it's dumb simple.
Notebook in thread:
How does it work?
The process is super simple. We simply query each model individually:
- Claude 3 Opus for reasoning + personality
- GPT-4 for reasoning
- PPLX for freshness/up-to-date info
Then, Claude combines the strengths of each and responds with a final, ideal output.
Apr 5 • 6 tweets • 2 min read
Introducing `claude-researcher` 📈
A powerful Claude 3 research agent that delivers thorough reports in record time.
Just provide an topic, and a chain of AIs with **access to Google** will generate an incredibly comprehensive report for you.
And it's open-source!
`claude-researcher` is a constrained agent -- meaning its behavior is highly-controlled, leading to better results than open-ended agents.
It chains together lots of Claude 3 calls (and Google access) that work together to create a detailed report on a topic of your choice.
Apr 3 • 7 tweets • 2 min read
Introducing `Claude-Author` 📕✍️
One prompt -> an entire novel!
Just describe the high-level details, and a chain of AI systems will write an entire book for you in minutes.
- complete w/ cover art
- packages your book as a real e-book
And it's open-source!
Previous AI book-writing systems produced mildly interesting books that were filled with errors and quite boring.
Claude-Author is the first AI system that actually produces readable books.
Still not perfect, but it's a leaps and bounds improvement over previous approaches.
Mar 27 • 5 tweets • 1 min read
Introducing `claude-llm-trainer` ✍️
The world's simplest way to train a task-specific LLM.
Just write a sentence describing the model you want.
A chain of AI systems will generate a dataset and train a model for you.
And it's open-source.
How it works:
- The user describes the model they want
Ex: "A model that writes Python functions"
- claude-llm-trainer leverages a chain of Claude 3 calls to create a great dataset for your task.
- We process the dataset, and train a LLaMA model!
Mar 25 • 5 tweets • 2 min read
Introducing `claude-journalist` ✍️
The first Claude 3 journalist agent.
Just provide a topic, and it will:
- Search the web for articles/real-time details
- Choose the best sources and read through them
- Write a fantastic, *factual* article + edit it
And it's open-source!
If you want to try it, you can head to the Github repo in the last tweet in this thread.
But if you don't want to bother with code, I've built an even better + FASTER version into HyperWrite -- try it here: app.hyperwriteai.com/personalassist…
Mar 22 • 6 tweets • 2 min read
Introducing `claude-investor` 📈
The first Claude 3 investment analyst agent.
Just provide an industry, and it will:
- Find financial data/news for key companies
- Analyze sentiment/trends for each
- Rank stocks by investment potential + price targets
And it's open-source!
`claude-investor` is a constrained agent -- meaning its behavior is highly-controlled, leading to better results than open-ended agents.
It chains together lots of Claude 3 calls that work together to analyze the major stocks in a given category.
Mar 21 • 7 tweets • 2 min read
Introducing `claude-opus-to-haiku` ✍️
Get the quality of Claude 3 Opus, at a fraction of the cost and latency.
Give one example of your task, and Claude 3 Opus will teach Haiku (60x cheaper!!) how to do the task perfectly.
Just describe a task, and a chain of AIs will:
- Generate many possible prompts
- Test them in a ranked tournament
- Return the best one
And it's open-source: github.com/mshumer/gpt-pr…
`claude-prompt-engineer` is a constrained agent -- meaning its behavior is highly-controlled, leading to better results than open-ended agents.
It chains together lots of Claude 3 calls that work together to find the best possible prompt.
Oct 19, 2023 • 7 tweets • 3 min read
Introducing the world's most powerful AI Assistant.
Personal Assistant is NOT just another AI chatbot.
It can:
- Operate your browser to actually complete tasks
- Cite sources, so you can trust what it says
- And so much more.
You won't believe what Personal Assistant can do:
Personal Assistant combines everything we've built to create the single most capable Assistant on the planet — from researching, to carrying out tasks for you, and much more.
For example, here is the Assistant writing a well-researched marketing email AND sending it!
Sep 12, 2023 • 8 tweets • 3 min read
Here's a simple guide to set up your OpenAI Playground for day-to-day use, as a (better!) replacement for ChatGPT.
I've been getting so many questions about this, so hopefully this is helpful!
First, why would you want to use the Playground over ChatGPT?
- Greater system prompt/behavior control
- Save multiple system prompts
- Temperature/creativity control
- Longer outputs for reasoning prompts/working with longer text
- Non-nerfed models :)
- Edit all messages
Etc.
Aug 23, 2023 • 5 tweets • 2 min read
This is the world's simplest way to fine-tune a task-specific GPT-3.5.
**Just write a sentence describing the model you want.**
A chain of AI systems will generate a dataset and train a model for you.
gpt-llm-trainer is a constrained agent -- meaning its behavior is highly-controlled, leading to better results than open-ended agents.
It chains together lots of GPT-4 calls that work together to create a great dataset for you.
Aug 16, 2023 • 5 tweets • 2 min read
Introducing `gpt-oracle-trainer` ✍️
The easiest way to create a chatbot that can answer questions about your product.
Just paste in your product's docs, and a chain of AI systems will generate a dataset and train a LLaMA 2 for you.
And it's open-source: github.com/mshumer/gpt-or…
gpt-oracle-trainer is a constrained agent -- meaning its behavior is highly-controlled, leading to better results than open-ended agents.
It chains together lots of GPT calls that work together to create a great dataset for you.
Aug 9, 2023 • 6 tweets • 2 min read
Introducing `gpt-llm-trainer` ✍️
The world's simplest way to train a task-specific LLM.
**Just write a sentence describing the model you want.**
A chain of AI systems will generate a dataset and train a model for you.
And it's open-source: https://t.co/ANXr0SXPOjgithub.com/mshumer/gpt-ll…
gpt-llm-trainer is a constrained agent -- meaning its behavior is highly-controlled, leading to better results than open-ended agents.
It chains together lots of GPT-4 calls that work together to create a great dataset for you.
Aug 2, 2023 • 7 tweets • 2 min read
Introducing `Agent-1`: a breakthrough foundation model that can operate software like a human.
This is the brain powering Personal Assistant.
We’re already well above previous state-of-the-art, and we’re improving massively each week.
More details:
First, why are we building this?
Current hosted APIs are amazing — but operating software isn’t a task today’s models can handle reliably.
Even the next generation of unreleased closed models aren’t up to the task (and trust me, we’ve tried).
Jul 11, 2023 • 6 tweets • 2 min read
Introducing `gpt-prompt-engineer-classify`✍️
An agent that creates optimal GPT classification prompts.
Just describe the task, and an AI agent will:
- Generate many prompts
- Test them in a tournament
- Return the best prompt
And it's open-source: github.com/mshumer/gpt-pr…
This is part of the larger `gpt-prompt-engineer` project I open-sourced last week.
Now, you can use it to do more than create generative prompts -- with this update, powerful classifiers can be created automatically.
Just describe the task, and a chain of AI systems will:
- Generate many possible prompts
- Test them in a ranked tournament
- Return the best prompt
And it's open-source: https://t.co/rcnlJ5g5ZNgithub.com/mshumer/gpt-pr…
gpt-prompt-engineer is a constrained agent -- meaning its behavior is highly-controlled, leading to better results than open-ended agents.
It chains together lots of GPT-4 and GPT-3.5-Turbo calls that work together to find the best possible prompt.
Jun 27, 2023 • 7 tweets • 2 min read
Introducing `GPT Personal Assistant` -- now available to everyone!
AI that can operate your browser to complete nearly any task.
It makes *very* silly mistakes and is not yet reliable.
May 12, 2023 • 8 tweets • 5 min read
Comparing @Google Bard, @AnthropicAI Claude, and @OpenAI GPT-4 performance on tricky prompts:
Note -- I did all these as zero-shot, first-try prompts with default model settings.
May 10, 2023 • 4 tweets • 2 min read
I spent a few minutes improving the GPT-4-32K product manager prompt.
Now, it makes a table with the frequency of issues (numbers are too low, but the prioritization is correct), and thinks through what we should do next to improve the product.
System prompt:
You are the world's best product manager.
User prompt:
```
{FEEDBACK HERE}
```
Above is feedback from our users who have uninstalled our Chrome extension. I want you to first create a table with the main problems our users state and the frequency of each.
May 6, 2023 • 4 tweets • 1 min read
The leaked 'Google and OpenAI don't have a moat' memo has stirred up debate.
Some see the future as open-source, others believe in closed models.
My view: the future lies in the middle.
Open models continually catch up with existing closed capabilities but lag in cutting-edge:
Why?
Back in 2020-22, GPT-3 excelled in writing & coding, while open-source LLMs struggled.
Now, in 2023, GPT-4 handles more complex agentic tasks well. Open-source LLMs don't.
5 subscribed
