syntaxing

7 ms·

91.
by syntaxing 4mo ago
This is simply not true and a misrepresentation of the issue. There is no issue with SP500 buying into all these companies. The issue is that “valuation” should be determined by the market before the index funds buy into it hence the origin
92.
by syntaxing 4mo ago
Moral issues aside, this is a pretty clever way to get a wide variety of training data
93.
by syntaxing 4mo ago
What about self host Chinese models?
94.
by syntaxing 4mo ago
Hot take, barring from special edge cases, I find using dumber models (like local Qwen 3.6) to be the best balance. Smart enough to do stuff but dumb enough where I don’t trust it and verify what it’s doing rather than letting it do the thi
95.
by syntaxing 4mo ago
But worse data retention and training policy. It’s a balance.
96.
by syntaxing 4mo ago
If you look at openrouter for Deepseek model providers that does not use your data to train, it’s still significantly cheaper than Anthropic pricing. The Pro and Flash performs closely to Opus 4.5 and Sonnet 4.6 respectively (though no visi
97.
by syntaxing 4mo ago
It’s wild. Regardless of Deepseek direct pricing, on Openrouter itself, the pricing for Pro is comparable to Haiku. Flash is even cheaper. You get Opus 4.5 and better than Sonnet 4.6 performance.
98.
by syntaxing 4mo ago
Serious question, I have a Bambulab printer now. What is a good next printer option? Prusa is way too expensive without a decent AMS alternative. Flashforge is doing some sketchy stuff with their maker site + AI. Is Creality or QIDI the bes
99.
by syntaxing 4mo ago
Any brand of test kit you recommend?
100.
by syntaxing 5mo ago
Is there something that is good to be a “android” server? I want to sign in to this server for all my chat stuff and use beeper to connect to it. I tried using a tablet but the battery keeps dying.
101.
by syntaxing 5mo ago
Nope. My family was below the poverty line at one point.
102.
by syntaxing 5mo ago
Hot take, I truly believe the answer for HS is yes. I grew up in a school district where the teachers had to have a master degree in the subject they were teaching. Those teachers strongly shaped me who I am today and I believe their advanc
103.
by syntaxing 5mo ago
What really bothers me with the data center boom is that prior to AI, water and electricity was constrained in a lot ways here in the states. “Our infrastructure cannot support electrification of cars and trucks!” “No watering your lawn or
104.
by syntaxing 5mo ago
Wow can’t believe the voice engineer lead for Nabu Casa is here! Super excited to see if this works for HA!
105.
by syntaxing 5mo ago
This would be amazing for home assistant.
106.
by syntaxing 5mo ago
Competition is always good. I got a Mac Neo recently to supplement my larger 16” MBP and they really nailed it. It’s the perfect laptop for kids and travel. Most importantly it feels like it’ll last for a decade like my MBP. I hope it’s the
107.
by syntaxing 5mo ago
Funny how fast people forget. LAN mode was NOT part of their original plan until outrage like this happened last time. They shifted their course and changed their blog post after. Putting pressure as a customer is how you steer company’s di
108.
by syntaxing 5mo ago
You have LAN mode only because everyone was up in arms the first time. LAN mode was not part of the plan at first and Bambulab was forced to offer after “listening to their customer”.
109.
by syntaxing 5mo ago
I’m so torn about bambulabs. Prusa needs to redesign their core one so it doesn’t use 3D printed parts (I get how that’s part of their philosophy but it’s not working anymore), $400 cheaper, and have a reliable AMS system. There’s just no o
110.
by syntaxing 5mo ago
How has opencode go been for you? Worth changing over from Claude pro?
111.
by syntaxing 5mo ago
Yeah, openmanet with reticulum seems the most “professional” right now
112.
by syntaxing 5mo ago
I’m guessing it’s just haloW without the licensing requirements.
113.
by syntaxing 5mo ago
It sucks how everything feels like a toy. I think meshtastic is the closest thing to a “product”. They made a bunch of bad architectural decisions that are haunting them now like how nodes broadcast its info.
114.
by syntaxing 5mo ago
I know it’s all open source and I’m not paying for anything so I cant be choosy. But after playing with a bunch of Lora peer to peer chat systems. All I wish is a chat service that uses haloW. Since it uses wifi backend, regular wifi should
115.
by syntaxing 5mo ago
Dad and millennial here and this change has been very noticeable in my circle of friends including myself and I’m all for it. Men have been doing their share of housework too. But I will say, it’s not all dads but enough that I think this w
116.
by syntaxing 5mo ago
I always wondered if Justin Kan’s Atrium closed door prematurely by just 2-3 years. It would have been cool to see a “technology” driven law firm and how it would have adjusted to LLMs.
117.
by syntaxing 5mo ago
With turbo quant, you would reduce it by over 6X.
118.
by syntaxing 5mo ago
This is a very interesting strategy that might pay off. This model is a very good option for enterprise self host. I would argue a lot of companies are VRAM constrained rather than compute constrained. You could fit 4-5 running instances on
119.
by syntaxing 5mo ago
Is RAG dead? I would be very surprised a local small SOTA embedded model like llama-embed-nemotron-8b doesnt outperform the Haiku layer for this application. Should be pretty cheap and easy to prove out. With 32K context size, you can liter
120.
by syntaxing 5mo ago
Which quants are you using? I had similar issue until I used Unsloth’s. I would recommend at least UD_6. Also, make sure your context length is above 65K. https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF

More ›