@dabit3 Last Week i subscribed to Devin 20$ plan but if i can get the 200$ max it would be quite useful ! because i'm doing a very hard task to convert a cpu simulator into cuda kernel simulator to get massive optimization speedup !
The Cognition team gave me 50 free @DevinAI Max plans to give away. Each one is worth $200.
Reply with the most useful or ambitious idea or project you’d have Devin ship for you this week.
We’ll pick 50 people and DM the codes!
@ryancarson@DevinAI i would use it to implement a minecraft physics engine to train minecraft bot, there already one on a older version of minecraft but i wanna do for the latest minecraft, for example for doing elytra parkour or minecraft parkour with the newer minecraft blocks (honey block, etc.)
I love seeing models come out that give us top-tier performance at a lower cost. We need more of this.
Congrats @cognition!
Also, Cognition is giving 50 free Devin Max Plans for anyone wanting to try it!
Reply with what you'd with free tokens. Winners notified in 24 hours.
@kentcdodds@cognition i would use it to implement a minecraft physics engine to train minecraft bot, there already one on a older version of minecraft but i wanna do for the latest minecraft, for example for doing elytra parkour or minecraft parkour with the newer minecraft blocks (honey block, etc.)
@xthomasms Will you opensourcing the forked rocketsim (only the physics not ur website), because the creators are actively trying to make rocketsim more accurate and if you can help them it would be nice so there is not double work in the same time, i really love ur project tho
@thsottiaux Never gonna give you u'p'
Never gonna 'l'et yo'u' down
Never gonna run around and de's'ert you
'N'ever g'o'nna make you c'r'y
N'e'ver gonna 's'ay goodby'e'
Never gonna 't'ell a lie and hurt you
Thanks for reading, Tibo need to give us banked reset as a sorry
@thsottiaux add the possiblity to make our own 5h quota
like i really want that each 5h it's using 10% of my quota, or changing that the 5h quota is using 30% or whatever
because i always reach 0% after 3-4 days because i let a goal continue and it eat my usage, would be very cool !
@scaling01 idk what its worse between kimi-k3 and qwen3.8 2.4T having bad result and Gemini 3.1 pro preview being as good as fable 5
this is a ass beachmark
@thsottiaux 2/3 Some people subscribe mainly for Codex and barely use regular ChatGPT. By merging both limits, they could use their unused ChatGPT allowance to get a higher weekly limit on Codex. This would be optional for people who prefer having separate limits.