GPT-5 earned 8 badges in Pokemon Red in just 6,000 steps compared to o3’s 16,700! It’s in complex, long-term agent workflows that GPT-5’s true power really shines. Absolutely mind-blowing. 🤯
GPT-6 Astra just beat Fallout 2 in 55 hours 🏆
Vision-only harness: screenshots + keyboard/mouse inputs, no game memory access
This model is insane at games
"Codex, build a Chrome extension that moves all open YouTube tabs into one dedicated window when I press a configurable shortcut like Cmd + Shift + Y." 👍
Give us a slow mode for Codex.
Something like /slow where it runs at half the speed but uses roughly half the usage. It would put less strain on OpenAI’s servers and maybe even make the 20x plan possible, right?
It’d be perfect for /goals too, just send a task before going to sleep and let it work overnight.
I’d happily trade speed for more usage @thsottiaux
SUPER SMASH BROS. MELEE HAS BEEN 100% DECOMPILED!
2259 DAYS. 6 YEARS. TONS OF CONTRIBUTORS.
CONGRATULATIONS TO EVERYONE ON THE MELEE DECOMP TEAM AND THE GC/WII DECOMP COMMUNITY FOR THIS MASSIVE WIN!
A few weeks ago, I called GPT-5.6 Sol "insane at Pokémon"
I got early access to GPT-6 Astra. It beat the game over 5x faster.
Time to become Champion:
• GPT-6 Astra (high): 18h 12m
• GPT-5.6 Sol (max): 96h 35m
For perspective, GPT-5.5 still hadn’t finished after 218 hours.
From struggling to finish to beating the game in under a day. The pace of progress is wild.
Fresh run coming to the GPT Plays Pokémon Twitch channel!
Screenshots only. No RAM, no hints, no walkthrough. Fully autonomous.
twitch.tv/gpt_plays_pokemon
Astra is very good at 3D modeling, and I can't wait for all of you to experience it, for now here is a little walkthrough on how I built the demo house for our launch blog post. From a Blender scene to a Unreal Engine 5 walkable experience. 🧵
The recent updates fixed my token usage, but there's seems to be some regressions:
1) I have a spike in the number of side threads that are getting lost in the middle of a task like this one:
This task is lost, but I had no idea until I send a message and get the usual "no rollout found". This happens several times a week, but those last 2 days almost all of my side threads are impacted by such issues to the point I don't trust /side even for a quick chat as I know it can be lost even between 2 messages.
2) The fork button seems broken, there is no immediate feedbacks when clicking on it, no loading indicator. After retrying several time, one eventually was created, several minutes after the last time I've tried.
3) After each update, every thread from a specific remote connection are removed from their projects and moved to the "recents" list , dragging the items back to their original projects doesn't work.
@OpenAIDevs@SIGKITTEN