Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Introducing MIRA. A playable, multiplayer world model. A dream of Rocket League. Trained on 10k hours of data collected with publicly available bots, MIRA learns the dynamics of a four-player game. The model runs in real time at 20 fps, based on the keys you and the other players...

507,564 Aufrufe • vor 3 Monaten •via X (Twitter)

35 Kommentare

Profilbild von kache
kachevor 3 Monaten

@ricklamers You probably don't want to predict the entire world

Profilbild von General Intuition
General Intuitionvor 3 Monaten

Play the four-player demo now.

Profilbild von Palmer
Palmervor 3 Monaten

10k hours sounds like a lot then I remember some people got like 5x that by themselves lmao

Profilbild von Vincent Micheli
Vincent Michelivor 3 Monaten

Come hang out with us at our ICML booth to play some matches and learn more about MIRA!

Profilbild von 0xobsessed
0xobsessedvor 3 Monaten

i don’t understand what you’re building

Profilbild von paula wehmeyer
paula wehmeyervor 3 Monaten

Play MIRA at ICML in Seoul with us. Meet the team at booth 111.

Profilbild von Aakash Reddy
Aakash Reddyvor 3 Monaten

Here I was thinking MOBA games would be the first to have a world model

Profilbild von Eddy Xu
Eddy Xuvor 3 Monaten

woahh

Profilbild von Mark Grinev
Mark Grinevvor 3 Monaten

unless world models lead to net new creativity in gaming or a harry potter off the dome esk design why am i trying to play a knock version of an already existing game

Profilbild von Garry Fan
Garry Fanvor 3 Monaten

I'm being griefed by my teammate pls add a report button 9/10

Profilbild von Ronak Malde
Ronak Maldevor 3 Monaten

If AI progress stalled here I wouldn’t be too upset

Profilbild von justin
justinvor 3 Monaten

wtf

Profilbild von signüll
signüllvor 3 Monaten

this is incredibly cool.

Profilbild von Garry Fan
Garry Fanvor 3 Monaten

6 in the morning and I'm playing Rocket League in my browser holy shit this is cool!

Profilbild von Joseph Jojoe
Joseph Jojoevor 3 Monaten

this is insanely cool

Profilbild von rohit
rohitvor 3 Monaten

the most impressive demo so far in 2026 - it's actually playable

Profilbild von Rahul Karajgikar
Rahul Karajgikarvor 3 Monaten

insanely cool stuff, as a rl player this is very exciting

Profilbild von josh halliday
josh hallidayvor 3 Monaten

i am going to come and destroy MIRA at rocket league 🤣🤣

Profilbild von James Xu
James Xuvor 3 Monaten

this is insane 🚀 super cool work

Profilbild von thom
thomvor 3 Monaten

God we are taking off so hard

Profilbild von offmylawn
offmylawnvor 3 Monaten

what the actual

Profilbild von Chris Klein
Chris Kleinvor 3 Monaten

Multiplayer world models before GTA 6.

Profilbild von acai
acaivor 3 Monaten

@0xm1kr @doppelfun

Profilbild von Alvaro L
Alvaro Lvor 3 Monaten

Fascinating

Profilbild von Ethan Goodhart
Ethan Goodhartvor 3 Monaten

insane

Profilbild von Seth  Phillips
Seth Phillipsvor 3 Monaten

Our friends just played it, absolutely incredible work.

Profilbild von venivci
venivcivor 3 Monaten

oh.

Profilbild von AL Robin
AL Robinvor 3 Monaten

The name is the same as the AI agents inside Telegram by The Open Platform :)

Profilbild von yang
yangvor 3 Monaten

Curious where medal data was used in this process since it says it was trained off bot data exclusively. “We train on roughly 10,000 match-hours of recorded 2v2 Rocket League matches, generated entirely by self-play between game-playing bots; no human gameplay at all. “

Profilbild von Team Reagent
Team Reagentvor 3 Monaten

@FelionSpike Wanna try it out? I absolutely called the fact that we would replace entire games with world models with a persistence mechanism

Profilbild von Rey Neill
Rey Neillvor 3 Monaten

Latency is rough, but impressive work!

Profilbild von Griz
Grizvor 3 Monaten

watching a model dream the next frame straight from your inputs with no engine underneath is the part that gets me

Profilbild von Neheart
Neheartvor 3 Monaten

10k hours just from bot-generated data is wild, curious how it handles real player unpredictability versus what the bots taught it.

Profilbild von Apurva Mishra
Apurva Mishravor 3 Monaten

This is sooooooo coooool!!!!!

Profilbild von Sebastian Buzdugan
Sebastian Buzduganvor 3 Monaten

20 fps matters, but recovery from bad player inputs decides if it feels playable

Ähnliche Videos

At Avalon we are building "Real-time creating" - the ability to generate gameplay ready persistent worlds prompted from text. While others are building real-time video world models, Avalon is building real-time world generation inside a fully playable, persistent multiplayer engine. Internally running at 3840×2180 at 60 FPS. Built on Unreal Engine. Multiplayer by default. Persistent by default. Gameplay-ready by default. This is not a video latent replay. Not a simulation of interaction. It is a real 3D world with physics, logic, and authoritative multiplayer state. Avalon is trained on proprietary Avalon interaction data and powered by a hybrid system that combines language understanding, 3D model generation, procedural systems, and structured gameplay logic synthesis. Players can walk through a live world and generate environments, assets, mechanics, and entirely new gameplay modes using natural language. We accomplish this through a combination of 3D model generation, game logic generation based on our proprietary systems, and AI driven world creation. While other players are inside it. Changes persist instantly. State is synchronized in real time. Creation happens inside the world, not outside of it. Describe a biome. Spawn a civilization. Create a survival mode. Build a dungeon crawler. Launch a new game inside the world. Avalon interprets intent and integrates it directly into the live multiplayer environment. This is not a world model predicting video. This is a gameplay engine that understands language. If you can describe it, you can build it. And others can walk into it instantly.

AVALON

66,037 Aufrufe • vor 7 Monaten