Havoc

10 ms·

271.
by Havoc 3mo ago
There is also uruky. They’re functionally unlimited on $5 but less developed feature wise
272.
by Havoc 3mo ago
One more walled garden enshitifying rapidly
273.
by Havoc 3mo ago
Check at the local thieves guild eh White House I mean
274.
by Havoc 3mo ago
Now watch the octogenarian UK lawmakers outlaw them anyway...right after they print the proposed law out because they don't do so well with computer stuff
275.
by Havoc 3mo ago
I've got worker nodes on m920qs and proxmox VMs too for workers, but like the clean segregation of a HA control node group. I doubt it matter much either way tbh. K8S is inherently resilient. I did have the rasps on hand though...not s
276.
by Havoc 3mo ago
I'm glad they're doing them. Audits are decidedly imperfect, but on balance people tend to toe the line better on good practices when they know they're being audited.
277.
by Havoc 3mo ago
>architectural thinking became way more important and difficult. Did it though? I think the architectural stuff was always there and was always hard. And still is. I don't buy that it got harder now that you've got a pretty sma
278.
by Havoc 3mo ago
Which I disagree with hence my comment?
279.
by Havoc 3mo ago
You may have just caught the wrong end of it timing wise. They used to be super stable and cheap but being a low cost provider they have no real buffer to absorb even minor hikes let alone the shitshow we saw in ram/ssd.
280.
by Havoc 3mo ago
Went a slightly different angle - Rasp 4s aren't impressing anyone these days, but 3 of them make for a fine HA controlplane. Once you've got a stable HA controlplane its much easier to connect whatever hardware you have lying aro
281.
by Havoc 3mo ago
It can be both at the same time - easier and different.
282.
by Havoc 3mo ago
>boy Seriously?
283.
by Havoc 3mo ago
>verbose GLM defaults to max effort btw https://docs.together.ai/docs/glm-5.2-quickstart#reasoning-e...
284.
by Havoc 3mo ago
Flash Lite: 0.3/m and 2.5/m Deepseek Pro: 0.435/m 0.87/m That's wildly ambitious pricing by Google. You can maybe get away with spicy pricing at the SOTA edge but at the lower tiers everything is a lot more price se
285.
by Havoc 3mo ago
Seems like a solid writeup. I like the point about open models being deaccel overall but helping with diffusion. That seems like a net win
286.
by Havoc 3mo ago
OAI feels far more cooked than anthropic. One of them heading to IPO and the other opting to not show their books should tell you everything
287.
by Havoc 3mo ago
Fun as the asic play would be it has a giant hole - context storage. Raw speed only gets you so far if you can’t store and cache
288.
by Havoc 3mo ago
>What makes the Chinese models this good? Why wouldn't it be? China is pumping out AI research and researchers at a staggering pace and there is no inherent reason why western models should be better
289.
by Havoc 3mo ago
oh wow - hadn't realized they decided to opensource Qwen 3.8 Max. That's pretty big news.
290.
by Havoc 3mo ago
Nice. I'll give that a go. Pretty vanilla config so fair chance it'll work
291.
by Havoc 3mo ago
I'd prefer it uncensored too but this objection was insightful in 2023. Everyone and their dog knows they're censored on topics sensitive to their country, much like western models are censored for western woke sensitivities. Nobo
292.
by Havoc 3mo ago
People saying copy aren't wrong...but if you can offer a copy at 1/5th of the price then you've got a winning product not a copy
293.
by Havoc 3mo ago
TIL I'm on 0.56 and hadn't realized there is a change. Thinking I'll wait as long as I can and then just get an LLM to translate current config to lua once the internet has been seeded a bit with examples
294.
by Havoc 3mo ago
Bonus discovery I just learned. Under nvtop -> F2 -> Processes -> Fields -> there you can make it show encode/decode. It's disabled by default
295.
by Havoc 3mo ago
No iGPU on my CPU (5800x3d). So was straight software decode (well maybe AVX2?) against GPU decode. GPU route was 80W more. joebonrichie's comment was excellent though - there is a new nvidia tweak that eliminates the GPU going to a hi
296.
by Havoc 3mo ago
Sorta. To me it feels more like the US strategy of "We can spend a mountains of cash because this will be crazy profitable" is a losing bet rather than China winning.
297.
by Havoc 3mo ago
Alas this is white label mem off eBay and fair bit of time has elapsed so not confident exchange would work. Minor miracle the ECC works at all given that At some point I should try running it at slightly lower speed i guess. It’s very hard
298.
by Havoc 3mo ago
Oh that’s interesting. I shall investigate thanks. edit: very quick test with MPV seems to confirm this works - eliminates the gnarly additional GPU draw. Measuring draw at wall now puts CPU and GPU route on equal footing...both basically t
299.
by Havoc 3mo ago
Or my personal favorite - offshore teams being confidently wrong after having asked AI
300.
by Havoc 3mo ago
Interesting - will see if I can toy with it. Word of caution though. I’ve found that on my machine (Linux/nvidia) unaccelerated video is way more power efficient. The second video is playing it kept the GPU in a high power state and th

More ›