linolevan

6 ms·

61.
by linolevan 8mo ago
What's funny is it's available on /v1/models, but if you call it you get an error saying it's not accessible yet. No word on pricing, probably the same as 4.7 if I had to guess (0.6/2.2)
62.
by linolevan 8mo ago
It does not have vision. On the Z.ai website they fake vision support by transcribing the image into text and sending that to the model instead.
63.
by linolevan 8mo ago
Spawned out of a bet with some of my coworkers, 100% generated by Claude in the web UI, scraped from realtimebillionaires.de
64.
How many of the 3,191 billionaires can you name?(billionaires.linolevan.com)
1 points by linolevan 8mo ago|1 comments
65.
Art of Roads in Games(sandboxspirit.com)
611 points by linolevan 8mo ago|207 comments
66.
by linolevan 8mo ago
Quick feedback: Website is basically unusable on mobile
67.
by linolevan 8mo ago
It’s pretty neat. Had some previous discussion that may be interesting on https://news.ycombinator.com/item?id=46595393
68.
by linolevan 8mo ago
You have the right intuition – AVIF is an image format based on the encoder for AV1 (which is a really, really good video codec). If you're interested in video, you might be interested to know that AV2 is in development.
69.
After self-hosting my email for twenty-three years I have thrown in the towel (2022)(cfenollosa.com)
5 points by linolevan 8mo ago|1 comments
70.
by linolevan 8mo ago
Really interesting! Genuinely a good puzzle, I wonder how frontier models do here.
71.
Theorizer: Turning Papers into Scientific Laws(allenai.org)
2 points by linolevan 8mo ago|0 comments
72.
by linolevan 8mo ago
This model is sort of interesting since it seems to be using a lot of synthetic training data – but your point stands
73.
Zotac warns component shortages threaten 'survival' of GPU manufacturers(tomshardware.com)
3 points by linolevan 8mo ago|0 comments
74.
by linolevan 8mo ago
I'm particularly excited to see a "true base" model to do research off of ( https://huggingface.co/arcee-ai/Trinity-Large-TrueBase ).
75.
Trinity large: An open 400B sparse MoE model(arcee.ai)
231 points by linolevan 8mo ago|82 comments
76.
by linolevan 8mo ago
Speculation on my part: Your site either supports accessible text scaling, or it doesn't. If only partly supports it – it might as well not at all.
77.
Try text scaling support in Chrome Canary(joshtumath.uk)
120 points by linolevan 8mo ago|47 comments
78.
by linolevan 8mo ago
For tiny models, the SFT data mixture is unbelievably critical to usability. They are unable to generalize in almost any way. If you don't have multi-turn conversations, they will not be able to do multi-turn conversations. If you have
79.
CSS selectors are global and evaluated RTL(bsky.app)
1 points by linolevan 8mo ago|0 comments
80.
by linolevan 9mo ago
I believe SmartOS is a distro of Illumos (in the same way that debian is a distro of linux).
81.
by linolevan 9mo ago
I'm not convinced that LLM training is at such a high energy use that it really matters in the big picture. You can train a (terrible) LLM on a laptop[1], and frankly that's less energy efficient than just training it on a rented
82.
by linolevan 9mo ago
Had a small discussion about this on an OP on bsky. A somewhat interesting discussion over there. https://bsky.app/profile/simonpcouch.com/post/3mcuf3eazzs2c
83.
When "likers'' go private: Engagement with reputationally risky content on X(arxiv.org)
55 points by linolevan 9mo ago|28 comments
84.
Electricity use of AI coding agents(simonpcouch.com)
119 points by linolevan 9mo ago|65 comments
85.
by linolevan 9mo ago
Not sure what the default is – whatever that was. It's probably the quantized mlx4 if I had to guess.
86.
by linolevan 9mo ago
Tried it within LMStudio on my m4 macbook pro – it feels dramatically worse than gpt-oss-20b. Of the two (code) prompts I've tried so far, it started spitting out invalid code and got stuck in a repeating loop for both. It's possi
87.
The <Geolocation> HTML Element(developer.chrome.com)
1 points by linolevan 9mo ago|1 comments
88.
by linolevan 9mo ago
This is a crazy change. I wonder if part of the reasoning is that sites without a robots.txt tend to be very low-quality. Search is a very hard problem and in a world of LLM-generated internet, it's become way harder.
89.
What came first: the CNAME or the A record?(blog.cloudflare.com)
466 points by linolevan 9mo ago|162 comments
90.
Flipping responsibility for jobs in SpiderMonkey(spidermonkey.dev)
2 points by linolevan 9mo ago|0 comments

More ›