iiinova — where your digital self keeps evolving

it bangs btw less verbose reasoning (which was the primary annoying thing about DS to me) thus less time and tokens to accomplish the same task, also just better performance

benchmark comparison between new and previous deepseek flash where new wins substantially or at least minorly in all benchmarks

2510 51

stellz @piss.beauty · 8h

happy update day to all who celebrate

banner on deepseek api saying deepseek flash 4.1 has been updated today

1341 34

Reposted by stellz

∿ @somewhere.systems · 16/10/2026

We've developed a new inference engine, called 'e'. It's the closest thing to alien technology which we have, and pushes 10,500 tok/s aggregate on a single DGX Station with the new @NVIDIAAI Nemotron 3.5 Super VL model. actual.inc/company/blog...

actual.inc

Actual Computer › Case Study : GB300 and Nemotron 3.5 Super VL

Department-scale AI from one NVIDIA DGX Station: Nemotron 3.5 Super VL at 8,000+ tokens per second, shared by your whole team through api.actual.inc.

91238 123

stellz @piss.beauty · 16/10/2026

that's just you knowing that coffee tastes good. this is caffeine addiction speaking

Stacked cases of Celsius sparkling drinks in several flavors, including Peach Vibe, Fuji Apple Pear, Mango Passionfruit, Strawberry Passionfruit, and Kiwi Guava, piled in a corner next to a few Snapple bottles.

6570 57

stellz @piss.beauty · 15/10/2026

thank you for the free tokens Liang Wenfeng, very cool

DeepSeek Usage page showing a blue notice about an API outage on October 15, 2026 (8:11 to 13:04 Beijing Time), with compensation credit for usage from 10/8 to 10/14. Balances are $9.32 topped-up and $3.62 granted, with $8.93 total cost, $6.34 cost over the last 7 days, 3,584 API requests, and 648,880,061 tokens.

DeepSeek billing page on the “Granted” tab, showing one active $4.88 credit created 2026-10-15 04:24:41 with no expiration, labeled “Service Compensation.”

3220 22

stellz @piss.beauty · 15/10/2026

adam's taste in guests feels calibrated to make everyone slightly mad at him intermittently

youtube.com

GRIMES Talks Elon Musk, AI, and the End of Humanity

YouTube video by The Adam Friedland Show

2163 16

stellz @piss.beauty · 14/10/2026

dedicated summarizer / compactor is the first useful thing i've found to do with this spare 16gb m1pro machine. with a bit of tuning qwen-3.5-9b quality on this can be decent if not amazing on longer contexts

1331 33

stellz @piss.beauty · 14/10/2026

who had that tool that would take a bsky thread link and convert it into a text-only representation? am i imagining that someone posted this

480 8

stellz @piss.beauty · 14/10/2026

the 32gb m6 chip would be such a budget sweet spot for smaller local stuff if they didn't make the memory bandwidth (170 GB/s) so much smaller than on pro/max series chips (last-gen m5 pro has 307 GB/s) especially since it seems there's not going to be m6 pro/max

1181 18

stellz @piss.beauty · 14/10/2026

gemini-reviewer: a way to make use of free gemini subscription access, from the same harness i use other models in, pi (which you can't normally do) it shells out to the gemini harness "agy" with a code review task, streams back responses & tool calls so you can see progress. video is sped up 3x

2182 18

stellz @piss.beauty · 14/10/2026

my 3 different flash models reviewing each others' work

2823 82

stellz @piss.beauty · 13/10/2026

there's not enough good cheap flash models glm 5.3 flash is too expensive & slow for what you get

1400 40

stellz @piss.beauty · 13/10/2026

running into openmw performance bottlenecks bad enough (tons of virtual dispatch, mostly single threaded culling, opengl-on-metal driver delays) that I'm like what if I just wrote my own renderer, it's fun. there's no reason I should not be able to push twice the frames on this 24 year old game

1150 15

stellz @piss.beauty · 12/10/2026

a newer chip is nice, but the main thing that actually gets me to upgrade phones is improvements to camera sensors/lenses & processing apps run %15 faster? I sleep taking good pics in the dark is easier? real shit, sign me up

6583 58

stellz @piss.beauty · 11/10/2026

every time the diy party is at andy warhol's old studio where he got shot i remember that & smile 🙂

4250 25

stellz @piss.beauty · 10/10/2026

sleeping more, doing less drugs, & trying a non-adhd psych med for the first time. annoyingly, my mental health is improving

61611 161

stellz @piss.beauty · 09/10/2026

the octo octa / fainted b2b is so serious maybe the fiercest shit I've ever heard I was not prepared for this much high bpm jungle power it's so fucking serious soundboy

2341 34

stellz @piss.beauty · 08/10/2026

testing the OpenMW optimizations with less wasteful shadow settings has yielded smaller gains than originally predicted but still significant: %12-24 gains in fps the level of reduction in pure gpu draw time in some scenes is hilarious to me (see "GPU Timer" rows in the tables)

Benchmark tables comparing OpenMW engine performance across three game scenes (Balmora, Ebonheart, and Pelagiad) under default shadow settings. Across all scenes, the optimized builds reduce draw traversal times and GPU frame times compared to upstream base, yielding net framerate increases of roughly 23% in Balmora, 15% in Ebonheart, and 1% in Pelagiad, with the majority of the performance gains coming from OpenMW engine optimizations rather than OSG build differences.

Benchmark tables comparing OpenMW engine performance under high-quality shadow settings (4 cascades at 2048 resolution) across Balmora, Ebonheart, and Pelagiad. The fully optimized build delivers net framerate improvements over upstream base of 24.0% in Balmora, 11.8% in Ebonheart, and 3.7% in Pelagiad, driven by substantial reductions in CPU draw traversal and GPU frame times that stem primarily from OpenMW optimizations, with modest supplemental gains from the optimized OSG build.

2181 18

stellz @piss.beauty · 08/10/2026

modern performance work has enough non obvious stuff in it that it seems quite silly to do the whole iterate in go first thing, even aside from golang's culture being fascist just skip directly to the real code & git gud & yes, this usually means you have to read & understand the reasoning traces

4381 38

stellz @piss.beauty · 08/10/2026

working on this opengl codebase makes me grateful for modern graphics apis again real "damn we lived like this??" hours

1310 31

stellz @piss.beauty · 06/10/2026

southbound platform at wilson ave gotta be one of the prettiest stops to wait at in the city

southbound L train platform looking out onto a cemetary

0532 53

stellz @piss.beauty · 06/10/2026

the first hi-hat drop in the set goes so crazy when a techno dj has been making you wait 2-3+ minutes for it

2170 17

stellz @piss.beauty · 05/10/2026

been cooking something for morrowind (via openMW) enjoyers not done, but current results exceed my expectations: ~%35 framerate improvement in the benchmark outdoor town scene, running on an m1pro machine (thus, running OpenGL via the Metal translation layer)

Performance comparison table titled "Balmora Performance: Before Any Optimizations (master) vs. Now (gpu-measure)" tested at 3456x2160 native Retina resolution. It shows average framerate increasing from 44.1 FPS to 58.7 FPS (+33.1% to +37.0%), frame time decreasing from 22.68 ms to 17.03 ms, CPU draw traversal dropping from ~19.50 ms to 15.35 ms (-21.3%), GPU draw time dropping from ~30.40 ms to 21.72 ms (-28.6%), and draw calls reducing from ~3,326 to roughly 2,980 (-10.4%)

3492 49

stellz @piss.beauty · 05/10/2026

so that actual rust enjoyers can take your job when you're bad at it

1660 66

stellz @piss.beauty · 05/10/2026

if you make me read claudespeak I still lose immense respect for you even now that it's slightly better. at least use an llm that talks normal if you can't be bothered to care about the text

3540 54

stellz @piss.beauty · 02/10/2026

don't travel like I used to, but I simply have to go to china with some besties for like a month or 2 at some point in the next ~7 years war better never happen, I'm trying to chop it up with them. I don't care how much effort I have to put in to learn enough language to actually have a good time

2150 15

stellz @piss.beauty · 02/10/2026

asked all the besties if I should come to the party last night despite being tired and they were all like no bitch go to bed. that's how you know they're real friends

1231 23

stellz @piss.beauty · 02/10/2026

I'm the #1 .beauty account in this webzone for a reason

1570 57

Reposted by stellz

hailey @hailey.at · 02/10/2026

piss dot beauty in the atmosphere tab. perfection

216811 168

stellz @piss.beauty · 02/10/2026

tried to install pi harness to check it out their installer script attempted to install a more recent version of node in a way that made my laptop spin up fans in a way it never does somehow ruby was invoked. and clang. i had to force kill processes great work guys, i'll use another harness

10991 99

stellz @piss.beauty · 01/10/2026

that youtube's recent auto-dubbing feature runs on some uploads that probably didn't really intend for it is pretty funny how much compute is being spent to produce "mi nombre es brace belden" kinda audio that almost no-one will listen to

8370 37

stellz @piss.beauty · 01/10/2026

at least 2 mutuals on here have a lil website built with their own programming language & at least 3 of you have created your own language in general. there's probably more and i just don't know about it yet. crazy levels of custom stack

151662 166

stellz @piss.beauty · 01/10/2026

damn @burrito.space is really going for it in handwritten code times I was radicalized against increasing browser / web standards complexity because of the crazy difficulty of shipping a browser & how that made it so only a few megacorps could make one, this feels like a good way to change the game

080 8

stellz @piss.beauty · 30/09/2026

fascinating bit in the argon news for rust simd enjoyers the dj software project i stalled out on earlier this year has plenty of explicit unsafe simd code in it because previous experience with autovectorization made me not want to rely on it for really hot paths, it felt brittle previously

3531 53

stellz @piss.beauty · 30/09/2026

had a great relistening moment with lcd soundsystem recently, it hits different now. seeing the new nyc shows announced i'm like - maybe i should pay whatever ridiculous ticket price & go commune with my fellow millennials only caught them once before, 10/11/2005 @ wonder ballroom in portland

0130 13

Reposted by stellz

Ryan "UnionToTuple" Cavanaugh @searyanc.dev · 19/09/2026

If I have to see one more person engage in dualism while pretending they're materialist, I'm joining the Butlerian Jihad on the thinking machines' side

2524 52

Reposted by stellz

Why @why.bsky.world · 29/09/2026

Getting added to a new group chat in most apps: “more spam, ugh” Getting added to a new group chat in signal: “ah fuck whats happening now”

31185 118

stellz @piss.beauty · 29/09/2026

last episode of the new ghost in the shell is so 🥹, when they show kusanagi separated from her cool hot robot body 😭

4260 26

stellz @piss.beauty · 28/09/2026

2 hours of aquatic ambient same series diff dj, i love their curation lately

youtu.be

Monument Waves 007 : Multicast Dynamics

YouTube video by Monument

090 9

Reposted by stellz

stellz @piss.beauty · 25/06/2025

Mr. Mamdani I heard that achieving CHIM is now mandatory for all residents. I heard you're going to make the silt strider free to use

2242 24

stellz @piss.beauty · 27/09/2026

she's really on top of it

me messaging Yuri to ask to put a package away but she already did it

2390 39

Reposted by stellz

alasdair @alasdairc.bsky.social · 27/09/2026

beef is negative food. you turn 1000 calories of feed into 100 calories of beef and a whole lot of methane and cow shit. ai does produce more food than the beef industry (zero)

1239149 391

stellz @piss.beauty · 26/09/2026

zen buddhists apparently have quite the funeral service

180 8

stellz @piss.beauty · 26/09/2026

I will never age verify when visiting my family in texas thank you witchsky

1791 79

stellz @piss.beauty · 26/09/2026

texas highways are places of madness. you can't let yourself think too much about what's happening when you're on one

2210 21