MIRA: Multiplayer Interactive World Models Trained on Rocket League
1–10 of 35 posts
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#2Václav here from the team, we're happy to answer questions :)
The most surprising part to me is the auto-recovery behavior we mention at the end of the blog post, since any other model I've seen always stays diverged once it goes off the rails once. But MIRA really doesn't like to be out-of-distribution. To be completely honest we're not entirely sure why this happens.
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#3it would have been easier to just go to FNAC and buy Rocket League like a normal person :)
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#4Wow! At first, I expected this to be a demonstration of an AI playing rocket league, but I rapidly realized this is actually a model simulating rocket league. Wild! It feels just like the real game.
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#5this is insane - what’s your thinking on how this improves model grounding and efficiency vs single pov outputs?
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#6Is this now the easiest way to play Rocket League on Linux?
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#7Is this now the easiest way to play Rocket League on Linux?
The checkpoint also weights half less than the game install! ;-)
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#8Nice to know my inability to play Rocket League with any level of skill carries over to this world model
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#9this is insane - what’s your thinking on how this improves model grounding and efficiency vs single pov outputs?
A lot actually, since the model has all information given to it in the four views, it doesn't have to deal with any "theory of mind" of modeling the other players or being consistent over long times. See [1], there's a video of a single-player model where a car disappears behind a ball and never reappears.
Multiplayer has its own risks such as the four views desynchronizing, but overall it gives a big boost to the model.
Re: MIRA: Multiplayer Interactive World Models Trained on Rocket League
#10We'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