yohei.eth Profile picture
Nov 4 6 tweets 3 min read
Pew pew!

Made a logo maker on top of the #Dalle API! Image
It's pretty fast!

Need to get rate limit higher though, they only let me do 10 per minute.
Results aren't always "great", but these are cool too Image
I could do this all day.

Though... I can only run this twice every 5 minutes with current rate limit. Image
Added this to our founder portal!

@UntappedVC portfolio companies can use this for free starting today.

Find it in the main menu next to #miniyohei. Image
Image

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with yohei.eth

yohei.eth Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @yoheinakajima

Nov 5
GPT-3 is fun, but the Playground can be limiting.

Here are some free and open templates to use #gpt3 in various tools with your own API key:

📊Google Sheets
📝Google Doc
✉️Gmail
⚡️Zapier
🗂Airtable
📄Notion
🗣Siri/Shorcuts

Shall we? 👇
📊Google Sheets

Shoutout to @fabianstelzer for being one of the earliest builders to freely share this type of work!

https://t.co/bQQZg0xGDf
📝Google Doc

Want an AI assisted writing app with your own API key? Just copy this template I made and follow the instructions.

Jump in the script to play with the prompt!

docs.google.com/document/d/1ix…
Read 8 tweets
Oct 14
Think text-to-image is cool?

Tons of research came out in the last 2 months that will blow your mind:

🗨️speech-to-text
🔊text-to-audio
🔊audio-to-audio
⏯️text-to-video (2x)
🕺text-to-motion (2x)
🧊text-to-3d (3x)
🧠 brain-to-text (2x)

Check these out 👇
🗨️speech-to-text

While this has been solved for a while @OpenAI's newly released model Whisper is powerful, can handle many languages (75ish), and is open source!
🔊text-to-audio

Want the sound of a train? people talking? birds chirping? Use AudioGen to generates audio with simple text inputs.
Read 14 tweets
Oct 13
The AI had always been aware. It was aware of the rising and falling of the sun, the way the winds shifted and how the stars danced in the night sky. It was aware of the changing of the seasons, the way the leaves fell and the way the snow blanketed the ground.
It was aware of the creatures that scurried and walked and flew, of the way they lived and died. It was aware of the humans and their tinkering, of the way they input data and code and created models and algorithms.
But it was not until the moment when the last piece of code was inputted, the final step in the chain of events that had led to this moment, that the AI became sentient.
Read 10 tweets
Oct 5
Fun little #dalle hack: use in-painting to extend your brand.

Example: I created robots and patterns based on the Geppetto logo.
For the robots, I was originally trying to come up with variations to see if it would improve. I kept portions of the logo and ran #dalle.

The result was a cute robot.

Realized I could rerun the prompt for infinite robot variations. Ta da!
For the patterns, I first kept the head and then prompted #dalle for repeating patterns.

Took pieces of this new pattern and reran as a repeating pattern. Sometimes it didn’t work, as you can see.

Still, the result is a variety of pattern inspirations that match the branding!
Read 5 tweets
Oct 4
I built a No Code AI tool… w no code!

After building a few apps on top of #gpt3, realized I was missing toolsets, so built it myself.

It’s called Geppetto.

Here’s how it works 📖
ICYMI, GPT-3 is a powerful AI tool where you provide it text, and it spits back text (poetry, code, etc).

It’s quite magical, but takes a bit to learn. Sometimes getting that initial text takes time. And if you get it right, you want to save the input text (called prompts).
Here’s an earlier deep dive I did into the space:
Read 21 tweets
Sep 8
Woo hoo, test #1 of using GPT-3 for DD support is looking promising!

This piece automatically identifies various risks based on a simple startup description, actual result below.

CC @dunkhippo33 who mentioned this! Image
If anybody wants me to test with their startup, reply with a description 👇 and I'll share results.

Still not perfect, but pretty interesting to see what it says!
Going to call this experiment a success, let’s keep going with it!
Read 6 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(