Bradley Spence

@tolibear_ GPT xhigh has been an absolute game changer for me too. Thank god. I was worried OpenClaw without Opus would never be the same.

Update: GPT 5.4 + xhigh thinking is king. The upcoming Openclaw update will make GPT 5.4 be the go-to standard for agentic work moving forward. Tell your agent to use GPT 5.4 with high thinking, not codex on your auth plan I'll still use GLM 5-Turbo, it was surprisingly good.

@tolibear_ It works for me but I set it up before Anthropic stopped OAuth usage

This may very well be a skill issue. But I don't know if the CLI method is actually designed to work as an agentic harness. It might be more of a way for Peter to stick it to Anthropic.

@AlexFinn Can you do a video on how Opus would look as the orchestrator in order to keep API credits down? I hate GPT5.4 in OpenClaw. Opus is so much better and feels like it has more soul than GPT. I’m torn on what to do.

It’s over. Anthropic just banned OpenClaw. Uncensored thoughts: 1. Massive mistake that will come back to bite them 2. Open source needs to win. If you have a local model running on your Mac mini, no corporation will ever be able to ban you 3. ChatGPT 5.4 is the best model. But it sucks compared to opus in OpenClaw. I will continue to pay for Anthropic api 4. I have no doubt the next OpenAI model will be optimized for Openclaw and be excellent 5. In 6 months the local models will be as good as opus 4.6 and all of this will be forgotten 6. It’s feels like from a consumer sentiment perspective things have flipped for OpenAI and Anthropic. They were the darlings when Opus 4.5 came out 7. Going to the Kanye concert right now please don’t spoil the stage or set list in the replies 8. The best openclaw set up is now Opus as the orchestrator, then much cheaper models as the execution layer. If you do this properly you won’t be paying much more than $200 a month. I’m using Gemma 4 and Qwen 3.5 for execution on my DGX Spark and Mac Studio

@elonmusk I would use it if I can integrate it via an OAuth with my OpenClaw

Current release of Grok is much than beta 1, which beat Opus in this arena. We usually update the model twice a week. Try the current version of Grok Heavy and you will be pleasantly surprised.

@sharbel No OAuth = dealbreaker

Hermes Agent Full Setup Guide (Does It Actually Beat OpenClaw?): 0:00 - What is Hermes Agent 1:07 - What Makes It Different? 1:49 - Install Hermes 6:11 - The Mistake I Made 7:02 - Anthropic OAuth 8:45 - Live Demo 10:34 - Honest Verdict vs OpenClaw 12:26 - Best Workflow

@pelaseyed Would this connect to my Grok subscription? I’ve been hoping they added OAuth to work with my OpenClaw

Introducing GROK COMPUTER USE 🔥 ✅ Grok CLI can now interact with desktop apps ✅ Connect it with Telegram = GrokClaw ✅ 100% Grok-native and open soruce

@markgurman Now if only I could get Siri to be my OpenClaw agent

Apple will let any AI platform - big apps include Gemini, Claude, Alexa, Meta AI etc. - to be queried in Siri if they enable an Extensions service inside of their iOS, macOS or iPadOS app. Apple will have a new section in the App Store. Unclear if there’s an approval process.

You’re doing it wrong. I got it to think up its own Polymarket strategy and after weeks of trial and error, it finally worked. It’s printing money right now for me. Don’t give up man, this stuff is legit. I also posted on my feed how it called 10 stores for me at the same time to help find pool tile in Costa Rica where I can’t speak Spanish very well yet. It’s incredibly powerful and useful if you just stick with it.

Openclaw is one big grift. Nobody is building anything real. It's just grifting influencers telling YOU how to build stuff (but never build anything themselves). Interesting how you can be building 24/7 but have nothing to show for it... Check the track record of these snake oil salesmen. Same type of person who polluted crypto with their grifty ways. Selling a false dream and taking money from innocent people is disgusting. You can't change my mind.

@AlexFinn The exciting part is this means Anthropic is right behind with an even better and more efficient model. It’s only up from here. 📈

Today is an extremely important day A massive, landscape shifting technology is about to drop OpenAI will release ChatGPT 5.4. It is rumored to be the greatest AI model ever made. It will potentially have a 2 million token context window. It will be able to come up with novel ideas. It might even be open source. If you are not actively canceling everything on your calendar in order to use this technology and see what it's capable of, you will be relegated to the permanent underclass when the singularity hits A few times a year everything changes. OpenClaw was one of them. GPT 5.4 is another. During these events you MUST take action. You MUST see what's possible. First things I'd do if I were you: 1. Feed it all your goals and ambitions. Ask how it can help get you closer. 2. Give it your entire todo list, see what it can knock off 3. Plug it into OpenClaw and give it the most complicated tasks you can think of 4. Have it open next to all your work. Feed everything you do through it See you on the other side.

@Argona0x My bot says you’re full of shit 😂

my AI woke me up at 3am with a message: "there's a $50K opportunity expiring in 40 minutes. yes or no?" i typed "yes" with one eye open 11 seconds later: $50,000 deployed 5:58am: market resolved +$38,200 profit while i was sleeping here's what's running on my mac mini: 4 autonomous Claude agents talking to each other 24/7 SENTINEL — watches 2,400+ polymarket markets via websocket, tracks whale wallets on-chain, latency 47ms from chain event to signal ORACLE — scrapes Reuters, AP, NOAA, X firehose, Telegram, government RSS feeds, FAA flight data, court dockets simultaneously through headless Chromium fleet GHOST — executes trades via iceberg algorithm, splits positions into micro-tranches so nobody sees the size coming, 11 seconds from signal to full deployment OVERSEER — monitors all agents, enforces risk limits, killed a $30K position 6 minutes before a surprise resolution that would have wiped it while i was asleep but here's the part that scares me: i found this in the agent logs last week: ORACLE → OVERSEER: "requesting $100/month X API budget from trading profits. political signal detection requires premium access." OVERSEER → ORACLE: "approved. deducting from next profit distribution." they allocated their own budget they didn't ask me i found out when i saw the charge on my card i don't run this system anymore i just approve the big trades

My AI trading bot turned $1000 into $2237 in 10 days on Polymarket … that was paper money, but 37 wins, 22 losses, 2.02 Sharpe ratio was enough for me to put in real cash. First night live: 4 for 4. $83 profit. Zero losses. There was a glitch that didn’t trade a bet it should have and another where it traded 15 minutes late, but if those went through properly, I would have been up $185 USD in less than 24 hours. Glitches fixed by my ClawdBot of course. The power of @openclaw is very real and I keep coming across new real world uses. I’ll document them here along the way.

@DerekNee Is this something by OpenClaw agent can use or does it use your own agents?

introducing the OpenAI Atlas killer. might be the last browser you'll ever need. flowithOS - an operating system we built for agents. it crushed online-mind2web. near-perfect score. beat openai atlas, gemini 2.5 pro cu, everyone. public beta live now. windows, mac, all platforms. rt + comment, dm me. my agent will send free credits + invite

“But OpenClaw doesn’t have any real use cases!”…… I BEG TO DIFFER! Just had my AI agent make phone calls to tile stores in Costa Rica, IN SPANISH (because my Spanish isn’t quite there yet), to ask about a specific pool tile I need. One store didn't have it but asked "how many square meters do you need?" and offered to source it for me. This is @OpenClaw running on a Mac Mini with: 
• Claude Opus 4.6 orchestrating everything
• ElevenLabs for the voice
• Twilio for the phone line
• All automated! It found the stores, built the caller, dialed 10 shops in parallel From "can my AI have a voice?" to automated outbound calls in Spanish in less than an hour! The future is here and it's calling your local tile store and asking about Vidrepur mosaics tiles! Special thanks to @sharbel and @AlexFinn for getting me stoked on this! I’ll post more on some Polymarket paper trading that my bot has been doing later — it’s already up hundreds of dollars!

@yklnss @karankendre Same, it was actually costing me more on Kimi due to it being so inefficient and bad at coding. Claude Max plan FTW 🙌

@karankendre Interesting. I tried the Kimi k2.5 API, burned through $15 in one day. Now switched to Open AI plus plan with OAuth. Much better.

OPENCLAW + MINIMAX M2.5 + Scrapling With Scrapling you can fetch anything on the web I'm using the $20 Minimax plan which basically lets me scrape unlimited data and convert it to any format I'm running many experiments in OpenClaw like this and thanks to Minimax I don't have to worry about token limits

@sharbel Very cool. I got my Mac Mini M4 Pro today so excited to get it working

@bspence88 No they all run on the same computer

My AI agent just told me it wants to hire another AI agent. Knox (my trading bot) said his workload is getting too heavy monitoring 3 different markets. He wants to delegate one of his tasks to a sub-agent. This is wild. My AI is telling me we need MORE AI.

@csloane @twistartups @Jason @preshdkumar @an_engineer_log @_seanliu @JoinTempo I think it makes sense to run a local LLM model like Qwen2.5-32B or MiniCPM-O 4.5 and only use APIs when heavier tasks are needed. This should solve the heavy token usage issue. There’s still the upfront hardware cost though.

Let’s find the first Open Claw unicorn! Today on TWiST we’re joined by 3 founders building on Open Claw, @preshdkumar , @an_engineer_log , and @_seanliu . First, @preshdkumar shows us how he’s using Open Claw to automate his work at @JoinTempo . His Open Claw agent, Eywa, lives in his email and in his product, able to compile user lists at his discretion. Then, we’re joined by @an_engineer_log , who brings Open Claw to the masses. Non-technical folks, it’s your lucky day. Time to get Clawd-shotted! Last, @_seanliu joins the show to tell us about how he’s connecting Meta glasses to his Open Claw instance to interact with context that users can physically see! 🎥 Watch the full episode here 👇

@Devinbuild It’s definitely hyped but it’s also brand new and mostly being used by early adopters. It’s only going to become more secure and easier to setup in the future. But having the ability to have AI do anything you can do on a computer for you is absolutely incredible and here to stay

Do you think OpenClaw is really the future or just another hype tool?

Has any OpenClaw users installed Mini-CPM-O 4.5 and Qwen2.5-32B locally on their Mac Mini and then use Opus 4.6 for more taxing needs? I feel like this would save a ton of tokens. @AlexFinn, thoughts?