vardalab

4 ms·

1.
by vardalab 12d ago
Well, if you are serious about it and you have Strix Halo, there are better ways of getting more context and capability and speed. Lookup halogen for Strix The most cost-effective local option right now, I think, is dual R9700. You can run
2.
by vardalab 13d ago
I don’t even think that regular Astra medium is anywhere close to 100 for tg
3.
by vardalab 18d ago
Just ask your freaking agent to read it for you and extract the information. That's what it's all about. Why would I be reading these articles other than information?
4.
by vardalab 19d ago
Who would you blame it on? Physics, gravity, destiny?
5.
by vardalab 21d ago
Technitium is also really good as a performant local blocker and recursive dns server. I have been running mine for years now and it is fast as well.
6.
by vardalab 24d ago
When I was doing research (physical electronics, lasers, fiberoptics and sensors stuff), lot of time was spent just writing all sorts of DAQ and processing code. So all this LLM stuff would have been really useful. There's a lot of dat
7.
by vardalab 25d ago
Yeah, it's like day and night. It used to be really unpleasant to interact with early codex versions. Even 5.3 wasn't great. Now, I go to Sol if I need to discuss anything. I don't even bother with Opus because I know that it
8.
by vardalab 25d ago
try using pi harness, hae not encountered these sort of problem myself also yuou can ask codex to look at the transcript and figure out the solutions to tool call failures that way
9.
by vardalab 25d ago
try ninfer once you get your 5090 https://github.com/Neroued/ninfer
10.
by vardalab 25d ago
Good 60-70 tg and 2K pp Qwen 3.8 27B FP8 can be had for about 5-6K (2xR9700 + PC) Gives about 3-4 concurrent sessions with full 262K Fast 150+ tg and 2-8K pp Qwen 3.8 27B nvfp4 is about 8K (5090 +PC) Gives really only one concurrent sessio
11.
by vardalab 25d ago
you can get 4xGB10 for <20K so that gets you about the same tg and pp will be probably better. Power consumption though will be something like 200W idle so that's a bummer. And you get VLLM and SGLANG unlike them mac where one has
12.
by vardalab 27d ago
Nice! I will need to get this on my mikrotik router.
13.
by vardalab 1mo ago
Using Sol XHigh or even High will deplete the Pro sub pretty fast in my experience if one is running any sort of automations in their harnesses. Sol-medium lets me squeek by with it using lesser subagents. Using ninfer on 5090 and 35BA3B q
14.
by vardalab 1mo ago
How would this work on the dual monitors? Is there an option to have a secondary monitor have just one workspace while main monitor has multiple workspaces?
15.
by vardalab 1mo ago
I actually changed the output style for Claude Code to use ASD-STE100 and it still doesn't help that much. It still comes up with a lot of stupid words like this gem "Standing where it stood"
16.
by vardalab 1mo ago
Opus yesterday produced this gem: "Standing where it stood." I replied, "never stop stopping!" and we had a stalemate, lol
17.
by vardalab 1mo ago
Opus yesterday produced this gem: "Standing where it stood." I replied, "never stop stopping!" and we had a stalemate, lol
18.
by vardalab 1mo ago
Yes, but we still have a leed in freedom according to the lore from our fearless leadership.
19.
by vardalab 1mo ago
That will be 1100 bucks, lol
20.
by vardalab 2mo ago
Sol 5.6 loves seams. I never heard the word mentioned until Sol 5.6 before that it never came up. Now everything is the seam, lol. At least it doesn't bear any load.
21.
by vardalab 2mo ago
So, I grew up in the former Soviet Union. Starting with ninth grade entrance entrance and exit exams were like that. I remember I had to take university physics exam just to be able to qualify for a consideration. It was fairly rough, but I
22.
by vardalab 2mo ago
Yeah, I told it to save in its memory that I don't want to have any more word salad!
23.
by vardalab 2mo ago
Says who? I kind of like it.
24.
by vardalab 2mo ago
It's at least 2.5x that speed for dual sparks and prefill is good as well. Basically going on vibes it is faster seeming than what one gets by default with openAI or Anthropic.
25.
by vardalab 2mo ago
My takeaway from this was that Claude was being completely useless as a helper Thanks to all the safety safeguards and that nonsense.
26.
by vardalab 2mo ago
Yeah, but maybe Kimi doesn't pretend to know better than I do. Like, I just had a task today where I had a silly form where I wanted to copy in a signature from one document to another document. And, you know, I could have just used so
27.
by vardalab 2mo ago
One thing I find that's missing a lot or at least I haven't come across other than commercial offerings like AquaVoice is a decent injected technical vocabulary so that the initial transcript requires minimum cleanup afterwards. B
28.
by vardalab 2mo ago
now it's back? i saw it was gone from the usage but then reappeared about 10 min later, lol
29.
by vardalab 3mo ago
A lot of stuff that has to do with VLLM and troubleshooting and compiling and building VLLM, compiling kernel, or just dealing with setting up eval for local models, it punts to Opus 4.8 on a regular basis. To the point that I have given up
30.
by vardalab 3mo ago
I don't know what your evals are, but you need to reevaluate them.

More ›