K0IN

3 ms·

1.
by K0IN 6d ago
I saw a lot of ppl think about what jev could use under the hood and could someone explain why this can't just be an embedding model where we just embed all the input + decisions and give back the cosine (or whatever) similarities?
2.
by K0IN 7d ago
My day job is improving data quality in big enterprises (we build products for this), and I see where this article is going, it's a bit watered down, but I see this every day at our customers, most companies just don't have a simp
3.
by K0IN 8d ago
Good video on the topic: https://www.youtube.com/watch?v=gjHh9g7yo9Y
4.
by K0IN 17d ago
now add a source tab and lets see, how many ppl. will fix it themselves and how many turns it needs.
5.
by K0IN 17d ago
I have used the flash model for over 3b tokens and ofc. I saw some hallucinations and premature termination (I also get this on Astra - way more often than with deepseek v4 flash), but I never had a infinite loop (using the copilot as harne
6.
by K0IN 23d ago
Am I missing something,I'm getting a 404?
7.
by K0IN 25d ago
upvote, k0ins hold together.
8.
by K0IN 27d ago
I just want to put this out there, smolmachines is a wonderful program to solve this, I use this mostly for stuff needing docker socket / docker in docker (example strix and agents). (I'm using podman on my host)
9.
by K0IN 29d ago
Me and my wife share calanders both ways between Google cal and iCloud cal its an absolutely shit show to setup, and I also rember having to use the Web app for certain settings how can we not have solved this by now?
10.
by K0IN 1mo ago
what a nice platform, i would love this to patrole my house and have some fun.
11.
by K0IN 1mo ago
first the automated part, that makes a pr "review ready", this helps us alot to gain confidence. 1. Automated tests that actually test real cases 2. Automated tests to test for regressions 3. Automated end-to-end tests, to make s
12.
by K0IN 1mo ago
I was also startled when python did ß.upper() returns "SS". Which is kind of unsuspected in some cases (if string length changes with an upper call)
13.
by K0IN 1mo ago
funny timing, i was exploring the same, some friends and me play alot of geogussr (a game where you put into streetview somewhere on the world, and have to find out where) lately. I want to make it more video game like, cause the jumps of g
14.
by K0IN 1mo ago
I think this is our third lap.
15.
by K0IN 1mo ago
you mean you dont want a singe company to have all your code?
16.
by K0IN 1mo ago
yeah but using the rule of thumb (I see floating in the Internet) which is sqrt(#params * #active params) which would give sqrt(284B*13B) = 60B, so deepseek should perform better.
17.
by K0IN 1mo ago
I used Qwen 3.6 27B extensively (>1B tokens) and DeepSeek V4 Flash (the older one also 2B+ tokens). And I just can't fathom that the new 3.8 beats the new DeepSeek V4 Flash (which, in my eyes, is one of the best everyday coding mode
18.
by K0IN 2mo ago
Some china proxies charge as little as 1 cent / Mio tokens for luna, which I use. As far as I know they bundle multiple codex subscriptions to get subsidized token pricings, so in codex it should be.
19.
by K0IN 2mo ago
As other calculated, even with a Mi300 you could not saturae it enough with one stream to break even with the DS API, so I think renting sparks would make it even harder cause they are considerably slower. (274gb/s vs 5.3tb/s)
20.
by K0IN 2mo ago
I think it's nice to have options, but I also don't see the appeal in "sub x mb" (except as codegolf or just for the hacky part, for day to day this does not matter) especially if features are missing or other compromise
21.
by K0IN 3mo ago
i use it too,can only recommend, also funny website btw.
22.
by K0IN 3mo ago
Sick that you can get 2 arm cores and 12 GB ram for free at Oracle cloud, did not know that
23.
Generating 3D assets with AI [video](youtube.com)
1 points by K0IN 3mo ago|1 comments
24.
by K0IN 3mo ago
Tldw: generating a general theme / color palette with gpt images -> using nano banana to split the images/generate single asset images -> creating 3d objects out of the images -> optimized the meshes using blender mcp
25.
by K0IN 3mo ago
To make it lazy. It's the match anything in-between so you can put stuff before and after its my most used regex. "my name is (.*?)$" => my name is k0in Or values "last (.*?), was great" => last sunday , wa
26.
by K0IN 3mo ago
Qwen 3.6 27b @ 90k context on a rtx 5090 (vllm)
27.
by K0IN 3mo ago
So my favorite regex (.*?) works? Puh.
28.
by K0IN 3mo ago
Is the classification model available, would be cool to have this in the toolbox.
29.
by K0IN 3mo ago
Awesome, I wanted to do the exact same thing (used gpt 5.5 + code) but it didn't get the model to work in onnx...
30.
by K0IN 3mo ago
with this size we could have a interaactive web demo.

More ›