I Trained an AI to Beat Trackmania’s Most Hunted World Record



I trained an AI using reinforcement learning to beat one of the craziest world records in Trackmania (A12)

Support the channel on Patreon:

Contact:
– Email: yoshtm.yt@gmail.com
– Discord: yosh_tm

Sound design and mix by Samuel Badel:

Music:
– Also Sprach Zarathustra – Richard Strauss
– Sonata No. 14 In C Sharp Minor, Op. 27 No. 2 ‘Moonlight’_ I. Adagio Sostenuto – Ludwig van Beethoven
– One to Save Us All – Bonnie Grace
– Ventus – Christoffer Moe Ditlevsen
– Rise of the Velcro – Gabriel Lewis
– Horisont – Owen Meyers
– Sumerian Paradise – Dew Of Light
– Orthosie – Ben Elson
– Swan Lake Dance of the Little Swans Act II – Pyotr Ilyich Tchaikovsky
– Dianna Beach – Dew Of Light
– Hiding Your Reality – Kevin MacLeod
– Gargoyle – Eden Avery
– The Flux Beneath It All – ELFL
– Waters Will Flow Again – Gabriel Lewis

source

25 Comments

  1. while some are making fun of the project of being useless…

    gamers are wandering what beefy PC was able to handle all those Ai's…

    & also some nation's military is weaponizing this to there mini drones & RC cars driven by Ai for Ww3 combat & spy & even signal guiding strikes… :X

  2. 2.351mil * average 30 sec (?) = around 60-80Mil seconds perhaps?
    At least 700 to 900 days of compute? Obviously parallel agents and multiple sessions and multiple hardware racks.. And these are huge assumptions there.

    Like, I don't know what kind of hardware was used, what AI model, the exact training process and decisions it took to do it, and how much time this added to the tally, but man, this is some electrical bill.
    The guy is a legend, I'd say go check if you are subbed.

Leave a Reply

Your email address will not be published. Required fields are marked *

You might like

© 2026 Cantinho do Vídeo - WordPress Video Theme by WPEnjoy