Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. MiMo v2.6(xiaomi.com)
    371comments
  2. Can gzip be a language model?(nathan.rs)
    2comments
  3. Spymarks, Not Watermarks(brand.io)
    85comments
  4. Transformers Explained Visually(poloclub.github.io)
    57comments
  5. Attention is all you have(alicegg.tech)
    219comments
  6. MiMo-v2.6-Pro: Intelligence, Performance and Price Analysis(artificialanalysis.ai)
    7comments
  7. What Sun got wrong(dtrace.org)
    326comments
  8. I don't want to read what you didn't write(colinbreck.com)
    215comments
  9. AI coding has made CI a bottleneck, so we reworked ours to keep up(linear.app)
    225comments
  10. NASA’s Mars Sample Return mission is dead(science.org)
    309comments
  11. Looking forward to Git 2.56 – and 3.0(lwn.net)
    43comments
  12. Divide by depth for instant 3D(gabrieloc.com)
    23comments
  13. Engineering Memory: On learning to memorize first 100 digits of pi (2024)(gregorygundersen.com)
    1comments
  14. Apple Music to open concert venue in Battersea Power Station(bbc.com)
    51comments
  15. The Advisory Group on Mathematics and Artificial Intelligence(terrytao.wordpress.com)
    60comments
  16. Socrates vs. the Written Word (2011)(wondermark.com)
    15comments
  17. World Wide Words(worldwidewords.org)
    discuss
  18. PDF Forgeries Are Surprisingly Rare (2022)(gwern.net)
    26comments
  19. How do traffic signals work? (2019)(practical.engineering)
    57comments
  20. Claude Status – Elevated errors for multiple models(claude.com)
    74comments
  21. Python Workers are now generally available(cloudflare.com)
    36comments
  22. Frontier AI on Your Own Hardware(timdettmers.com)
    75comments
  23. HERMES radio enables voice and data communication over vast distances(ieee.org)
    54comments
  24. Grok 4.7(x.ai)
    474comments
  25. Turn off and restrict access to Apple Intelligence features on Mac(support.apple.com)
    192comments
  26. Apple Copland D11E4 Booting in the Browser(pagetable.com)
    37comments
  27. More floating point alternatives(wizardzines.com)
    20comments
  28. First Shader from Zero in Godot 4(gdquest.com)
    7comments
  29. Why does mathmain need an encrypted loader?(safedep.io)
    36comments
  30. Roboharm: Do frontier robot policies refuse unsafe instructions?(robocurve.org)
    22comments

Truman World

23 pointsby 1d agotrumanworld.live
25 comments
4h agoHN ↗

How does this work? I couldn't find an about page?

4h agoHN ↗

The models used are mentioned on the page; primarily it's H3 Max Director, Jev, and an agent to manage context and send updates to Director.

3h agoHN ↗

Yes, I don't want to know what the tools are. What...is it? What's the goal? How is one meant to interact with it?

2h agoHN ↗

It's not on the page, but the creator (@WillTheRapper_) said on Twitter:

This is our first experiment in building a persistent Social World Engine — a world that continues to exist, evolve, and remember even when nobody is watching.

The audience is meant to direct the show. There's a crypto token involved in paying for it. Make of that what you will.

Personally I think it's an invitation to a lawsuit. Back in the day people were running AI Spongebob on Twitch and that got killed for copyright infringement. And this isn't a cartoon, it's Jim Carrey.

2h agoHN ↗

It is vibecoded slop, not sure why people are taking this seriously here

2h agoHN ↗

Some people are looking at what it is, rather than how it was made.

2h agoHN ↗

As avaer already mentioned it seems to be a thin wrapper around H3 Max Director, which generates real-time video and takes “natural language directions” that are then relatively seamlessly incorporated into the continuous video stream.

It's like a Choose-Your-Own-Adventure dialed up to 11.

I played around with it briefly on Fal.ai generating a cartoon, but it’s more of a technical curiosity for the time being since I don’t have a Scrooge McDuck vault of gold coins.

4h agoHN ↗

This is really an interesting experiment.

4h agoHN ↗

Truman looked down the street where an explosion occurred, then a shadowy figure came out of the smoke, approached down the road, ran into truman and then past him and accosted an indian man in a turban. They threw blows until an identical clone of the indian in a turban tackled the shadowy man to the rear of the frame. The first Indian man was handed a white monster energy drink. He took a sip. He immediately tranformed into Truman with glowing blue eyes. Truman was visibly confused and started saying garbled words like "what did you givrm." "I can feel the mermegh"

Then hugged to death.

3h agoHN ↗

"Now I am become HN, the Destroyer of Websites."

4h agoHN ↗

The star of the show is H3 Max Director [1], which lets you get a pseudo realtime WebRTC video feed which you can control with JSON messages.

The model is quite impressive, though obviously it depends a lot on what you're doing. But you can literally just type whatever you want and it will reasonably happen a few seconds later. Logical continuity is not included; attempting to maintain continuity is probably the bulk of the work for this experiment.

If you want to try your own version, you can vibe code an interface on top of the fal API in a few minutes, it is coding agent friendly.

Keep in mind it's super expensive; per list pricing (at least on fal), it costs $5/min to run, and the minimum is 1 minute.

[1] https://fal.ai/models/minimax/h3-max/director

4h agoHN ↗

the upvoting will continue until bandwidth improves

4h agoHN ↗

Error proxying request to container: There are more than 4096 concurrent connections inbound to the container, surpassing the limit

3h agoHN ↗

Do you

A) remove container, use raw process

B) 4096 concurrent connections is plenty, start dropping them and filtering more aggressively

C) Slap that cloudflare thingy and call it a day (same error rate, but users get told they are robots so it's their fault)

D) Just restart the server, the problem is no longer there, anyways

2h agoHN ↗

Dunno how this page is implemented, but the only thing that can't be trivially infinitely scaled is the streaming; the correct solution is probably to farm that out to a service. LiveKit or something fatter. 4096 connections is not necessarily enough for a livestream of anything.

1h agoHN ↗

The correct answer is E, you run more containers

3h agoHN ↗

This is really something. Well done. This is the future, albeit a bit rudimentary, but you just made me realize a few things.

In 10 years, the world will be very different, in 20 - unrecognizable.

2h agoHN ↗

What even is this? Everyone for $5?

2h agoHN ↗

If you opened page and haven't seen video, the page malfunctioned, try reloading it.

On the page there is streamed real time video that is generated by ai. The setting is similar to the movie Truman show with Jim Carrey where he was main character whose life was streamed 24/7. With the limitations of current technology, it is inconsistent slop, the consistency is at most 5-10s, they probably just generate first and last frame of a 10s "scene" and make 2 videos and then stich them together. There is some monetization scheme where users can steer what's happening on screen by paid prompts.

2h agoHN ↗

Couldn’t even tell what was happening here, and the moment the page loaded it made a bunch of loud noise at me so I closed it immediately. Comments indicate it’s slop anyway.

God damn I wish titles would tell you even a little about what a page is

2h agoHN ↗

Carelessly vibe coded and crashed under light load from a HN hug of death. The lowest bar in reliability.

Complete slop this is.

2h agoHN ↗

Looks jenky and cannot handle the traffic. Neat idea having a public-facing AI-agent observability panel wrapped around participatory improv fiction