victorsup

Anyone else noticed a significant decrease in the speed of all GPT-4 services, like me?

Amazing!

The unconventional takeaways Luck is important, which may surprise you. What worked for me might not work for you. Every successful story is 99.99% irreplicable. Reaching the break-even point is incredibly satisfying!

My launch strategy in general: make splashes & word of mouth Here is a an overview of what happened:

Stage 1: Making posts on reddit → 10 signups/day

Stage 2: Multiple ProductHunt launches → 200 signups/day

Stage 3: Shout-out from angel influencers → 1000 signups/day

Stage 4: More organic posts from influencers → 2000 signups/day

You have to be more specific. Do you mean feeding mailing and documents a LLM to train your new model?

personal charisma

like 90%

and screens not bright enough under sun. (Speaking from real experience)

Who is Tim? Tim Cook?

man, I used dark mode for everything...

you are correct

No library, built it from scratch

It was two-hour work (with the help of AI, of course). I was inspired by Midjourney (credit to them). But, yeah, as a nerd, I do not know how to bullshit. So I thought it's easier (for me) just to put the demo video and a tagline. Didn't even bother to write a description.

I implemented a very simple one for now, nothing too complicated. Some users are already "complaining" they need smarter auto-positioning algorithm

a visual parser for chatgpt output

MapGPT.com/ai & NodeGPT.com/ai are both sold...

[weird and cool] that's the best compliment for any hacker! Thx mate, you just made my day

select a text card and use backspace/delete key

I will DM Tim Cook, and make sure ChatGPT-3D releases on the WWDC with the apple AR/VR device.

I bet there are many developers drinking diet coke :)

can you elaborate?

true

this is great! Sadly, conceptgpt.com cost $3,499 on godaddy. I already bought chatgpt2d.com, it costs me $12

terrific idea! will do that if this makes money, lol

Hey! Glad you're enjoying the Superus-101 playground ! The waiting time (ragard a single thread) is actually dependent on OpenAI, with avg wait around 8-15 sec , not much I can do there. But, concurrent threads? Cool idea, definitely worth exploring! Thanks!

with NDA, I think it's fine

Can you share your secret source in your prompt?

what's the difference between your product and MS VScode AI copilot?