TopgAGi

@yashhq_22 If Claude was so good then codex wouldn’t be finding issues in its code none stop when in auditing it on every phase I build.

Solo founders, which model are you using for vibe coding? - gpt 5.5 - opus 4.7 - gemini 3.1

@CuriosityonX No such a thing as an alien

If aliens requested a meeting with a sole individual to represent the human race, whom should we send?

Wow @openclaw with @OpenAI gpt 5.5 changed my life It’s not too talkative and gets straight to the point It just works So good that I got the 20x plan today lol

@elonmusk She said

I love giant machines

@BillyM2k Hi

why do i have 2.8 million followers now if you’re not a bot say hi so i can see my real number

@openclaw @clawsweeper I just use codex for that

ClawSweeper 0.2.0 🦞 The OpenClaw maintenance bot now handles the loop: issue → @clawsweeper fix/build → guarded PR → review → repair → re-review → automerge Still conservative. Much less manual. clawsweeper.bot

@alphafox @grok is this ai?

This is why bracing is important: 😨

@Raytargt Imagine going to hell for making fake naked OF

A college kid I know in Austin showed me his Stripe at a poker night last week. $43,000 last month. OnlyFans. I asked who the girl was. He pointed at his MacBook. She is four text files. 47 photos trained her face. 90 seconds of audio cloned her voice. 1,200 words made her a person. Total build cost: $70. First week: $4. Fourth week: $1,847. Last week alone: $11,400. His phone buzzed three times during the hand. +$15. +$22. +$8. A real OF model films 4 hours a day, splits her revenue 50% with an agency, pays $4,000 a month for a chat team. 1,247 subscribers. $1,433 a day. Maya replies in 6 seconds and never asks for a 50/50 split. He paid less to build Maya than he spent on textbooks last semester. Three other guys at the table are building theirs this weekend. You have a MacBook. You have $70. You have one free Sunday. Next poker night, someone shows his Stripe. The question is whose.

@ProvenialAIHub @AlexFinn @davemorin Nah

@AlexFinn @davemorin Why openclaw over Hermes? 2 sentences

@davemorin pure curiosity: is this any different than me signing into openclaw with my chatgpt oauth before? has anything changed?

@Matt_Pinner Please do this captcha

you are sitting next to Elon Musk, what is the first thing you're asking him?

@fallawanna @elonmusk It’s not your face woman

Al was trained on women without consent. It included non-consensual nudity and rape edited to look consensual so it's allowed to be posted in the first place. Then you create those generations and you post them in our face showing us exactly how they're using our faces and bodies and voices without consent. It's sexual harassment.

Grok Imagine tutorial made with Grok Imagine. These is all AI-generated!

@elonmusk These is? These are*

@TiffanyFong Goodnight fish head

goodnight my little brown boys

@TiffanyFong Plot twist: he doesn’t even have a wife

So this freak fabricated the whole story and called his wife a little Asian fish head with no cannons ?

@AlexFinn @petergyang Use codex to update and fix issues

@petergyang same experience. very frustrating

Is there a better way to update OpenClaw than do this? Shit breaks half the time when I tell my bot to update itself.

@petergyang Use codex It fixes all the issues for me

@GaskinBryon @openclaw Yours is dumb lol

rolled back for 2026.4.23 already. They got to do something about this: Local models from .8b to 9B fail, no change on the Ollama side, just moving from 2026.4.23, then problems at 2026.4.26, continued into 2026.4.29, rolled back to 2026.4.23, fixed. Up to 2023.4.23 small local models worked just fine for routine tasks at chat.

OpenClaw 2026.4.29 🦞 💬 Group chats feel much better now 📌 Follow-up commitments from context 🔐 Safer exec, pairing, and owner controls 🟩 NVIDIA provider + model catalogs ⚡ Faster startup + plugin/channel fixes Group chat finally feels agent-native. github.com/openclaw/openc…

@pouria3 @openclaw Next time tell codex to fix the issue for you

Took a deep breath and typed “Openclaw update” and of course everything I had spent days setup stopped working!! Lots of Peter cursing and a few hours to figure out. But it’s just all worth it because that superhuman feeling Openclaw gives you is addictive. I am excited to test the GPT harness

@quxiaoyin @NousResearch Hermes good Deepseek nah If you want a Chinese model, stick to Kimi k2.6

I can’t believe I stopped using Claude Code max and entirely use DeepSeek and Hermes. It’s so fast, so so fast, 3x faster for the same task. So cheap. I spent $5 last week and never need worry about being rate limited or usage hit limits very two hours. For most tasks it’s perfect enough.

@openclaw @QuixiAI @opencode its ok gpt 5.5 is better anyways

@QuixiAI @opencode Anthropic is blocking us again. Lots of other companies with great subscriptions work though!

do any agents work with Anthropic Oauth? I thought they said it is permitted now, but seems like it still doesn't work? @opencode @openclaw