Activity
Mon
Wed
Fri
Sun
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
What is this?
Less
More
Memberships
Shipping Skool
201 members • $99/month
70 contributions to Shipping Skool
12h •
3 starter Macs for Sale
Hi all — lucky timing on my end: my area (Philly suburbs) had a brief window of stock at Apple Stores, so last week I was able to grab a Mac Studio (M3 Ultra/96/1TB) and a refurbished Mac Mini (M4 / 24GB / 1TB) for myself that I have specific plans for. That leaves me with three Macs I've been using for OpenClaw/Hermes that I no longer need and could reclaim my desk if I get rid of. I know I could get more flipping these on eBay, but I'd rather sell to someone in this community who's been held back by the shortage and the crazy eBay prices. All three are in perfect working condition — can be picked up locally, or I can ship them domestically in their original boxes. - Mac Mini M2 (8/256) — $300 + shipping. I'm the original owner. Originally used as a home theater server (now moved to Synology). This was my first machine for testing OpenClaw before I got hooked. - Mac Mini M4 (16/256) — $635 + shipping (The price I paid including PA taxes; still ~30 days to add AppleCare). Bought this before I found the 24GB/1TB version and don't need both. - Mac Studio M1 Max (32/512) — $1,300 (what I paid picking it up locally on Facebook from a photographer I know). Been using it for about 3 weeks with zero issues. I bought this to run a local LLM — this one can run around 27B Q4 if you dedicate the machine to it. If you're serious and one of these would help you get started, just DM me. I want to pass these along to someone who'll put it to good use rather than flip for a quick $.
1 like • 4h
im inspecting the info to determine if interested. Very glad you shared with the class first to try and help us out.
2d •
Local Hosting your LLMs
Hi everyone, we’ve just wrapped up our first session on self-hosting models. If you weren't able to join us live, the recording is now available and provides a comprehensive blueprint for local hosting. Please feel free to reach out if you have any questions
0 likes • 21h
- If I have a DGX Spark with a solid Qwen model running, how many openclaw agents or sub-agents can I have utilize this device and model at the same time? Or do each wait in a line for their turn to perform a task? - If I ran Hermes and Openclaw, would it be best to have two DXG Sparks so that each agent session would have a dedicated local model to use for regular daily business operations and activity? - Would it be expected to experience high-quality project research and building of apps using a good Qwen local model at the same time as other agents and sub-agents needing to get tasks and operations done simultaneously without issues? ( Example: I am on OpenAI Codex to orchestrate, and I may have Hermes doing 24/7 market monitoring and taking actions when certain triggers are hit. I also may have OpenClaw doing content marketing research and building and monitoring marketing campaigns as needed. Is all of this going to run ok on a single DXG spark model running a solid Qwen model? ) - From what I was understanding, I would need multiple devices and models running to regularly run many agents with sub-agents for regular operations like the example, and would still have to share the time needed to even access the device and use the model to perform tasks on each agent/sub-agent task when needed?
🚀
5d •
GPT 5.5 today
Just spotted on openrouter. Endpoints will go live today Get ready for testing this should be a massive upgrade over 5.4
0 likes • 2d
Game Changer with 5.5 - https://chatgpt.com/codex/
1 like • 2d
@Dan Carroll Glad you got your project knocked out. Its really is a good model. I am running 5.5 medium in codex and that is the sweet spot there so far. I have not even tried PRO. If I have a struggling project I might need to try it out. Yes with API that is a expensive idea to even try. The codex plan does not offer pro I believe. With the codex desktop app and 5.5 together things seems to work amazing together and not sure I need PRO. I have some projects in the pipeline and can let you know if still consider it an improvment or not.
6d •
Kimi K2.6
I dumped Claude Sonnet this morning for Kimi K 2.6, and it didn't break anything. It is surprisingly good. We'll see if it goes another 24 hours. If it does, I'll report back here, but it's really good.
1 like • 5d
Yes would love to hear more about your test results. I tested it out recently also and can say I like it better than GLM 5.1. Overall it seems to have a flow and vibe of getting things done like Opus so I am liking it so far. Will have one main agent further test as I build some more stuff and do some real things with it. Its definitely easier on the wallet and seems to be a good Opus fill in or replacement, will see.
0 likes • 2d
its just not GPT 5.5 or Opus but its better than some others out there. I stopped messing with it also. Its a backup if needed is all at this time.
5d •
Anthropic wants to remove Claude Code from it’s $20 subscription??
It appears Anthropic is A/B testing Claude Code access by removing it from 2% of prosumer “Pro” plan signups to gauge reaction: https://www.wheresyoured.at/news-anthropic-removes-pro-cc/ In the parallel news, open models are killing it: https://www.buildfastwithai.com/blogs/qwen-3-6-plus-preview-review Qwen 3.6 benchmarks are reported to be on par with frontier models such as Sonnet 4.5. Is privacy & independence important to you? Is this the time you leverage local models?? Share your thoughts!!
1 like • 5d
Yes Anthropic causing more issues again. Yes, I am excited about Qwen and other local models but need the models and equipment for local to match what I need and be affordable. Privacy and independence are nice but I need what works and I can afford most. I do want to learn more and find good use cases to get some serious gear and tinker more.
1-10 of 70
@stephen-king-8167
Some experience in tech and software, previous web designer. Became ill looking to learn a new skill to help with medical and life while disabled.
Active 2h ago
Joined Mar 28, 2026
Powered by