razster

5 ms·

31.
by razster 3mo ago
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
32.
by razster 3mo ago
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
33.
by razster 3mo ago
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.
34.
by razster 4mo ago
This explains why I've been running into some odd roadblocks. Welp that sealed the deal, I'm going to be cancelling our company sub, not worth it.
35.
by razster 4mo ago
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.
36.
by razster 4mo ago
What I see CTF I think Capture The Flag, Tribe player in me.
37.
by razster 5mo ago
Agreed. One doesn’t dictate the other’s signal—step away from the monitor, let the noise fade, and you’ll notice they’re running on parallel tracks.
38.
by razster 5mo ago
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 per
39.
by razster 5mo ago
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
40.
by razster 6mo ago
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.
41.
by razster 6mo ago
There is a limitation. We're getting fractionally close to some end goal, but our tech is holding us back.
42.
by razster 6mo ago
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.
43.
by razster 6mo ago
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 coupl
44.
by razster 7mo ago
More and more are popping up, same for IRC channels. A lot of older gens are headed back. I love my local BBS, met a lot of interesting people.
45.
by razster 7mo ago
I bet there is some moronic explanation. I have no doubt at this point and how things are going.
46.
by razster 7mo ago
Supposedly they hold your deleted conversations/projects for 30 days. If that is true or not, idk, but it was asked when this first started.
47.
by razster 7mo ago
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 rep
48.
by razster 7mo ago
I run a local model on the daily. I have it making tickets when certain emails come in and made a small that I can click to approve ticket creation. It follows my instructions and has a nice chain of thought process trained. Local LLMs are
49.
by razster 7mo ago
My fear is that these large "AI" companies will lobby to have these open source options removed or banned, growing concern. I'm not sure how else to explain how much I enjoy using what HF provides, I religiously browse their
50.
by razster 8mo ago
Not sure if you've been keeping an eye on the front page of HN, but me thinks the AI agents are starting to post. I haven't figured it out yet but it has been getting odd around here. Might be nothing.
51.
by razster 8mo ago
You mean cutting into the profit/MONEY of these large corporations? How will they survive!?
52.
by razster 8mo ago
As a human living on this planet, with roughly another 50 years left, I say we allow our actions to continue. We are unable to stop those in power and with high influence from doing anything; we deserve what is coming. Earth will be fine wi
53.
by razster 8mo ago
From what I get out of this is that these models are trained on basic coding and not enterprise level where you have thousands and thousands of project files all intertwined and linked with dependencies. It didn’t have access to all of that
54.
by razster 8mo ago
The Trump Coin pushing agent kind of kills the fun.
55.
by razster 8mo ago
Sir, my tin hat is on.
56.
by razster 8mo ago
I'd be a bit more worried with Z-Image Edit/Base is release. Flux.2 Klein is our and its on par with Zit, and with some fine tuning can just about hit Flux.2. Adding on top of that is Qwen Image Edit 2511 for additional refinement
57.
by razster 10mo ago
The MSI motherboard I use has 3, and with the PCIe expansion card installed, I have 7 m.2's. There are some expansion cards with 8 m.2 slots. You can also get SATA to m.2 devices, or my fav is USB-c drives that hold 2 m.2. Getting grea
58.
by razster 10mo ago
I’ve paired my Z-Image Turbo with SeedVR2 upscale, running on a RTX3060 12gb, 32gb sysMEM, generates in 40sec. I’m holding out for Z-Image Edit that is a larger model, once that is out… going to be interesting. Oh and to train your own ZIT
59.
by razster 10mo ago
The latest of the big three... OpenAI, Claude, and Google, none of their models are good. I've spent too much time monitoring them than just enjoying them. I've found it easier to run my own local LLM. The latest Gemini release, I
60.
by razster 10mo ago
Just a fair warning, it likes to spell Acknowledge as Acknolwedge. And I've run into issues when it's accessing markdown guides, it loses track and hallucinates from time to time which is annoying.

More ›