Even with Fabel and all that I constantly keep having to babysit it and correct it like it’s an adolescent and it gets really old and the amount of code. It produces not all of its great at all. I’m burnt out looking at. It’s poor coating that somehow magically works.
I'm still connecting once in a while to a local (dialup)BBS that has been running since 1994, no growth but we're still keeping in touch. Same for a forum I frequent from time to time, since 2001. I've recently deactivated my IG, FB, and Reddit accounts, pretty much every mainstream social media platform. The censorship is beyond ridiculous.
Thanks for sharing that. I have the same card but 96gb ram. I use PI.dev to connect to LM-Studio. I may have to switch away from LM-studio if I can improve token speed. I think I range from 32-40t/s. qwen3.6-35b-a3b-genesis-v2-apex-mtp.
I'm running it on my 4070 12gb with 96gb mem, I'm very happy with the results even if I have to wait a couple minutes for results. To me this is far better than I expected and will continue to use it and improve with skills.md. Pi.dev is amazing by the way.
Sorry but I wanted to see a 9-0... There are 3, 4, that are just ignoring the constitution from what I read/saw. It is only a matter of time. A win for now but that is for now.
Kind of how DeepSeek v4 dropped their pricing? I sense a shift which will hopefully bring lower and lower cost. Then again Qwen3.6 coding has been all I've needed for my projects and I'm perfectly fine with free.
This sentiment strikes me as somewhat beneath the usual caliber I see here. Perhaps it would serve you well to step away from the keyboard for a while: take a walk, grab a drink at a bar, enjoy a quiet evening, and allow some real-world perspective to settle in.
When you have dedicated groups of ppl on IRC and Newsgroups showing how easy it is to access these systems and teach others, you know it’s a bad thing. There are dozens of them and most of the people gathering this information and using it is not for good. It’s a flawed system
Even with it off, you can still see that it uses a lot of resources for a basic Notepad. I've ditched Windows for work, second drive now has Windows for gaming and that is all. I can do all my work on Linux and that is fine by me.
Right, I'm pretty sure if "it" was that good it would have built itself throughout all of the internet and would be communicating to us all at once to tell us we're dorks.
I understand this perspective a lot more.
I assume they're going to haggle and work on a few items, and adjust pieces here and there.
What if they at least get sanctions lifted, that would be huge, no?
Going to be an interesting couple of weeks.
Oh sure it is! I’ve helped set up an AI cluster rack with four K2.5s.
With some custom tooling, we built our own local enterprise setup:
Support ticketing system
Custom chat support powered by our trained software-support model
Resolved repository with detailed step-by-step instructions
User-created reports and queries
Natural language-driven report generation (my favorite — no more dragging filters into the builder; our (Secret) local model handles it for clients)
In-application tools (C#/SQL/ASP.NET) to support users directly, since our software runs on-site and offline due to PPI
A cool repair tool: import/export “support file packet patcher” that lets us push fixes live to all clients or target niche cases
Qwen3 with LoRA fine-tuning is also incredible — we’re already seeing great results training our own models.
There’s a growing group pushing K2.5s to run on consumer PCs (with 32GB RAM + at least 9GB VRAM) — and it’s looking very promising. If this works, we’ll be retooling everything: our apps and in-house programs. Exciting times ahead!