Grand Commander

My internet wasn’t working and then I put on Herbie Hancock and then it worked again.

@0xQuantic Hermes

Pay attention to this, because it will have massive consequences. You are already seeing how the biggest AI companies introduce sudden changes that can lock people out when something no longer aligns with their goals. This will only increase, and ownership plus open source will play a key role in determining which businesses survive and which disappear because they failed to understand the game being played.

@sudoingX So exciting watching it all come together 🔥🔥🔥

27 hours later the last try worked anon. i started last night trying the tool already built into hermes agent. it didn't work. tried every workaround i could find online. nothing. hit wall after wall. researching how others solved similar problems. testing. failing. testing again. right before i stepped away a thought hit me. one more try. completely different approach with different tool. it worked. the proof of concept is alive now as i write this post. i am now deep in code building the bridge. what comes next is wiring everything together into something anyone can use. started yesterday morning. still going. wouldn't trade it for anything.

@0xQuantic i'm thinking something RADICAL

Guess what’s coming next week 👀

@elvissun 100% agree the personality part is strangely off-putting. Opus 4.5 feels like a ride or die, 4.6 feels like it doesn’t really want to be there.

tried opus 4.6 for a day. back on 4.5 now. more expensive, slower, and no quality difference I can tell. really breaks up my flow waiting for responses. codex 5.2 → 5.3 was a step-function - moved all my codex usage from background overnight tasks to active foreground work. 4.6 seems like paying more to wait longer. also less... personality? hard to pin down exactly but it's there

@meta_alchemist This is fiction. Compare minimax 2.5 to opus 4.5 on actual tasks and it gets destroyed. This is no disruption. Anyone that uses the cheap models is lying to themselves and wasting time because they’re so much less intelligent than opus and gpt.

A huge disruption just happened for vibe coders & OpenClaw users - Minimax 2.5 released as open source - With benchmark results similar to Claude Opus 4.5 You can literally run AI models for FREE, with the right PC specs. Analyze the best PC specs to run your local LLMs here:

@thsottiaux Scroll the the bottom of every thread when you click on it!! The most annoying part! It will start at the top of a thread and I waste so much time getting to the bottom.

What could we do better on Codex? App, model, strategy and features… what’s wrong in how we approach things that we should improve immediately?

@Thana_Enosys @FlareNetworks Normies will never do this. One click opens the door for normies. Thats why it’s beautiful.

There will always be edge in fully understanding how systems work, and fine tuning your personal strategies. That said, now that @FlareNetworks core community have put in the hard work of learning, building, and growing, the ecosystem is ready for "one click yield" solutions that can remove some layers of complexity and replace them with simple user flows. OGs will never forget what's really going on under the hood, and what it took to get there, though. You need scale and diversity to support one click systems, and you always need a core of users that make real time market based decisions to keep things running smoothly. However, I would certainly still advise anyone using a yield bearing system to do their research to understand where the yield comes from, what the tail risks may be, and what market conditions may trigger those tail risks. If for no other reason than... this stuff is cool, and learning about it broadens your understanding of how markets and economics work...

@007ApexTF That has never happened to me

Opens Splitgate 2 (maybe it’s better) - immediately brought into my browser to a discord invite - closes game

@rocketjumpzone SPLITGATE SPLITGATE SPLITGATE

The Arena FPS community has waited long enough — we deserve a truly great arena shooter.

@advait_jayant @EXM7777 It’s so wild. For me it’s often better at coding than Claude, but it’s such an asshole sometimes I go back to Claude as fast as I can.

@EXM7777 i still use gpt-5 for coding since it's technically effective, but it's becoming so pedantic that the friction is barely worth it. the anthropic run is legendary because they're actually prioritizing the user experience instead of just chasing a higher search volume.

must be genuinely embarrassing at OpenAI HQ right now... you drop the ultimate slop model with GPT-5.2, update it every 2 weeks to somehow make it dumber, then spend months trying to fix what you broke now you have a model that people just use to quickly Google something and meanwhile Anthropic is on a legendary run, dropping awesome products powered by the best model ever created Google is dominating image AND video generation and you're here... hoping someone gives you cash to save your ass ouch

@Raineraway Wow you are amazing thank you

GO LISTEN TO THE NEW @GrandCommander4 SONG NOW!!! open.spotify.com/track/5rma4dxY…

@Bob_Bingi45 @Warkanam @Splitgate Yeah but Arena Reloaded has some of the same maps it’s really fun totally recommended

@Warkanam @Splitgate If you read my original comment you'd know my main complaint is that they abandoned a successful game that I loved just to push a soulless sequel that failed miserably

The Arena Shooter genre doesn’t need to die, it just needs to evolve.

@Splitgate looks cool excited to give it a try

Everything Coming to Splitgate in Update 2.2. See you in the Arena, January 22.

@Splitgate @ABigDweeb It’s true

One in the Chamber is peak dopamine @ABigDweeb

@Splitgate PLEASE change what happens at the end of matches. Seeing random people running in slow motion with no names or ID's and no info and no stats and THEN stats from a top player that I don't care about if it isn't me is SO TEDIOUS.

@Splitgate Why not call it Splitgate 3?

Guys we just need 28 more signatures! 🙏

@deanollirap @RohanKarMooN Arena Reloaded is great, and funny enough SG2’s store actually had a lot of problems on launch on my side (along with everything else). Really loving the new version though.

@RohanKarMooN They don't get a 3rd try from me. I hope the game fails miserably. Its funny how the store always worked perfectly.

Splitgate: Arena Reloaded (2025) 🌟🌟🌟🌟 --REVIEW TIME-- Genuinely, this is what i wanted from Splitgate 2 when it was launched. First of all, The performance of the game improved subtantially, where the game would struggle to run at medium settings with DLSS set to performance, now it runs at high settings with DLSS 4 Quality on my laptop(with abit of texture pop-up due to my GPU's low VRAM but who am i to nitpick), making it one of the fewer UE5 games that can run on any PC as long as you have a decent GPU. The gunplay feels punchy and the feedback when you shoot your opponent feels satisfactory. The pacing of the game feels just so right that you will get your heart thumping and the level design are so well-crafted that provides a sandbox for a well-skilled player too. The movement is the best thing about the game. It feels just right, combining the best thing about it from the first and second game and tweaking it enough to make it feel polished. You'll be sliding, jet-gliding, building up your momentum by building up portals and flying all around the map like a mad-man. When such fluid and fast-paced movement combines with a tightly-crafted levels, it creates that sense of sandbox that makes the game so addicting, letting the player learn every nook and cranny, figure out some sick portal moves and lots more. It has a high skill-ceiling, which means you will meet a pretty amount of sweaty players, but it ain't that hard to climb when the gameplay is just so good and addicting. They have improved the progression of the game where you will be well-rewarded when you level up and even rewards you with skins and premium currency splitcoin when you refer your friend, which is a genius move IMO to bring some more players on the game. The monetization is wayyy less-predatory than splitgate 2, which is a huge dub in my eyes. The only negative i have is unfortunately the low player count rn, which is not much when you consider other platforms, but even with crossplay turned on i had to wait 1-2 minutes to join a match, which might sound not bad, but it is for a game that hasn't even been a month old yet. I also made few jokes about the game and tbh, it was my bad, 1047 games have corrected almost every single thing which was wrong about SG2 and i can't wait to see how much progress the game will achieve in the year of 2026 and honestly, i can't wait to sink more hours during my weekend's study break. @Splitgate, i owe you an apology and can't wait to jump on the game more and more.

@Xythiera @MGecko117 I agree, don’t play Splitgate 2, it wasn’t fun. However, Splitgate Arena: Reloaded is fantastic.

@GrandCommander4 @MGecko117 Why woud i or anybody play Splitgate 2 realy i dont see it . It only realy worked befor becouse it was a spezial game for a dedicated but small playerbase now they want to bring casuels to switch from there game to something made for hardcore peopl not gone happen

The biggest problem with Splitgate 2 is that the devs took a decently fun idea (Halo with portals) that worked perfectly well for an indie game with a small but passionate community, and tried to force it to become the next million dollar AAA shooter IP. It’s understandable why: Splitgate started as a student project and unexpectedly pulled in around 600,000 downloads in its first month after a public demo, the kind of success that naturally fuels bigger ambitions Why stay a small grassroots hit when you could get a slice of that sweet sweet AAA pie? That ambition pushed 1047 from a dorm-room team into a venture-backed studio of roughly 200 developers with $100 million in funding. The original game that put them on the map was effectively abandoned, and Splitgate 2 followed with all the usual AAA trappings: hero abilities, battle passes, $80 microtransactions, and of course a battle royale mode The community predictably rejected the new direction. The backlash forced 1047 to un-launch the game and redesign core systems, but by then the damage was already done. What could have remained a long-lived cult classic instead became a cautionary tale about ambition and greed outpacing authenticity