HackerTrans
トップ新着トレンドコメント過去質問紹介求人

MasterScrat

2,633 カルマ登録 16 年前
General Intuition

florian dot laurent at gmail

投稿

Show HN: MIRA – Multiplayer World Model Trained on Rocket League

mira-wm.com
12 ポイント·投稿者 MasterScrat·4 日前·1 コメント

Meta's AI Storage Blueprint at Scale

engineering.fb.com
1 ポイント·投稿者 MasterScrat·6 日前·0 コメント

Machines do not press play: From temporal compression to model-ready tensors

spiraldb.com
1 ポイント·投稿者 MasterScrat·24 日前·0 コメント

The Origins of GPU Computing

cacm.acm.org
1 ポイント·投稿者 MasterScrat·3 か月前·0 コメント

Poolside hunts data centre partners after CoreWeave deal falls through

ft.com
1 ポイント·投稿者 MasterScrat·3 か月前·1 コメント

Keeping 20k GPUs Healthy

modal.com
2 ポイント·投稿者 MasterScrat·6 か月前·0 コメント

ClusterMAX 2.0: The Industry Standard GPU Cloud Rating System

newsletter.semianalysis.com
2 ポイント·投稿者 MasterScrat·8 か月前·0 コメント

General Intuition $134M seed to teach agents spatial reasoning using game clips

techcrunch.com
2 ポイント·投稿者 MasterScrat·9 か月前·0 コメント

Project Horizon: Why we're building a 2 gigawatt AI campus in Texas

poolside.ai
3 ポイント·投稿者 MasterScrat·9 か月前·1 コメント

コメント

MasterScrat
·10 時間前·議論
Playing backward is actually a great idea for testing generalization
MasterScrat
·3 日前·議論
Would be a great idea to see how much we could manipulate the latent space and whether it has some internal structure w.r.t the physical state. I guess the only unknown is how the world model would show robustness to latent states that are transformed through this network
MasterScrat
·3 日前·議論
Hey all, happy to see this here! This was a colab between General Intuition (that I’m part of), Kyutai and Epic Games.

You can read plenty of details in the blog post and tech report but the TLDR is that we trained a multiplayer world model on 10k hours of Rocket League data. We optimized it to be playable at 20fps on a single GPU.

So what you see in the demo is fully generated: there’s no graphics or physics engine. Instead it’s a 5b neural network that takes actions in and gives pixels out.
MasterScrat
·4 日前·議論
We're happy to release MIRA, a collaboration between General Intuition, Kyutai, and Epic Games.

Mira was trained on 10k hours of Rocket League data. The model has 5B parameters and runs 4-player games at 20 fps on a single B200 GPU.

We've released a playable online demo, an in-depth technical report as well as a 1k hour dataset of 4-players gameplay:

Technical report: https://mira-wm.com/paper Repo: https://github.com/mira-wm/mira
MasterScrat
·4 日前·議論
The checkpoint also weights half less than the game install! ;-)
MasterScrat
·14 日前·議論
This sounds like a strong statement with little backing. The author does infra at DeepSeek if his LinkedIn is to be trusted, and is the author of Foyer.
MasterScrat
·先月·議論
This is exactly what i would have needed 6 months ago
MasterScrat
·2 か月前·議論
I have mixed feelings (as in, I'm unsure how to feel) about projects where the code, the README and the HN/Reddit posts are mostly AI-generated.

I feel the frustration of reading "slop", but on the other hand the projects that surface do usually bring something useful to the table.

Should we simply judge the submission based on its technical merit? Why do I feel annoyed that an otherwise cool project uses typical LLM prose? For how long will we be able to recognize LLM-generated text, and what happens when we can't?
MasterScrat
·3 か月前·議論
and so does OpenAI
MasterScrat
·3 か月前·議論
Clearly they need more datacenters!
MasterScrat
·3 か月前·議論
Without paywall: https://archive.is/u7t8U
MasterScrat
·4 か月前·議論
> As we focus and compute demand grows, the Sora research team continues to focus on world simulation research to advance robotics that will help people solve real-world, physical tasks.

— https://www.businessinsider.com/openai-discontinues-sora-vid...

So yeah, focusing on world models
MasterScrat
·4 か月前·議論
This has a simple pragmatic solution though:

https://duckdb.org/2024/12/06/duckdb-tpch-sf100-on-mobile#a-...
MasterScrat
·4 か月前·議論
Are you saying Carmack isn't a "talented and successful developer"?
MasterScrat
·4 か月前·議論
Why would a game development pedigree correlate with rejecting AI? As Carmack said:

> AI tools will allow the best to reach even greater heights, while enabling smaller teams to accomplish more, and bring in some completely new creator demographics.
MasterScrat
·6 か月前·議論
> Have you ever noticed that the food graphics in Super Smash Bros. and Kirby Air Riders is flat “billboarded” stock images of food?

I have not... A video showing what it's talking about would have been a good addition to this article.
MasterScrat
·6 か月前·議論
Is there a reliable service that plots hourly price per GPU per cloud through time?
MasterScrat
·6 か月前·議論
Just so you know, this is already very much a thing on TikTok: AI-generated movie summaries with narrator voice explaining the plot while showing only major beats, reducing movie from 2h to shorts totaling 10min.

It’s honestly not the worse AI content out there! Lots of movies I wouldn’t consider watching but that I’m curious enough to see summarized (eg a movie where only the first title was good but two more were still published)
MasterScrat
·7 か月前·議論
Exactly - best case scenario is Reddit/HN front page with a cool project you enjoyed working on, have some nice conversations there, reach a few 100s stars which look good on your CV, and that’s it.

If you expect more long term support you better be paying me for my time.
MasterScrat
·7 か月前·議論
> Without world models. you cannot achieve reliability. And without reliability, profits are limited.

Surprising to simultaneously announce the end of the road yet point to the road ahead