#Compute

Blender Diariesblenderdiaries
2026-02-06
Dendrobatus AzureusDendrobatus_Azureus@bsd.cafe
2026-02-06

Fetching the video to learn

Thank you ๐Ÿ’• for sharing

youtube.com/watch?v=Ef70x0izkFU

@geerlingguy

๐Ÿฆ‹๐Ÿ’™โค๏ธ๐Ÿ’‹#Lobi ๐Ÿ’™๐Ÿ’•๐ŸŒน๐Ÿ’๐Ÿ’™๐Ÿฆ‹

#Raspberry #Pi5 #SBC #Compute #model #technology #notebook #Linux #OpenSource #programming #POSIX

Avinav Sahoo(he/ him/ his) (@avinavsahoo)

์ž‘์„ฑ์ž๋Š” '์šฐ๋ฆฌ๊ฐ€ ์‚ฌ์‹ค์ƒ ๊ณ„์‚ฐ ํšจ์œจ์„ฑ ๊ฒฝ๊ณ„(compute efficient frontier)๋ฅผ ๋„˜์—ˆ์„์ง€๋„ ๋ชจ๋ฅธ๋‹ค'๊ณ  ์ฃผ์žฅํ•˜๋ฉฐ Sam Altman ๋“ฑ ์—…๊ณ„ ์ฃผ์š” ์ธ๋ฌผ๋“ค์„ ํƒœ๊ทธํ•ด ์ค‘์š”ํ•œ ์—ฐ๊ตฌยท๊ณตํ•™์  ์ด์ •ํ‘œ ๊ฐ€๋Šฅ์„ฑ์„ ์ œ๊ธฐํ•ฉ๋‹ˆ๋‹ค. ์ด๋Š” ๋ชจ๋ธ ํ›ˆ๋ จ ๋น„์šฉยท์Šค์ผ€์ผ๋ง ํšจ์œจ ์ธก๋ฉด์—์„œ ํฐ ๋ณ€๊ณก์ ์ผ ์ˆ˜ ์žˆ์Œ์„ ์•”์‹œํ•˜๋ฉฐ AI ์—ฐ๊ตฌยท์ •์ฑ… ๋…ผ์˜์— ํŒŒ์žฅ์„ ์ค„ ๋งŒํ•œ ๋ฐœํ‘œ๋กœ ํ•ด์„๋ฉ๋‹ˆ๋‹ค.

x.com/avinavsahoo/status/20196

#compute #airesearch #efficiency #scaling

2026-02-03

I don't think rallying against AI no-matter-what is the right way to oppose its deployment.

I'm super bullish on Chinese AI labs, which, due to export controls favour efficiency in their models, and it is foreseeable that pushing for efficient AI models will allow us to break the massive centralisation and inevitable enshittification of AI.

Having skilled professionals wielding AI, similar to the philosophy of Primeagen's 99 AI coding plugin for neovim, will probably be a net benefit and augmentor rather than a detractor

Also, efficient, deployable at home models will even further eat into the negative margins that Western AI labs have, similar to how pushing against AI-derived work being copyrightable will make the billions invested by Hollywood into AI worthless.

Here I borrowed from @pluralistic 's thought that traditional copyright litigation is not a effective tool for protecting creators from gen AI, as that will primarily benefit major corporations rather than individual creators.

#AI #Compute #Copyright

Frank JonesFrankJones78
2026-02-02

NVIDIA Just Killed OpenAIโ€™s $100 Billion Deal

The landscape of AI infrastructure has shifted overnight. NVIDIAโ€™s latest move into direct cloud services has effectively undermined OpenAIโ€™s massive $100 billion compute deal, signaling a new era of competition between the hardware giants and the software pioneers.

technology-news-channel.com/nv

Don Curren ๐Ÿ‡จ๐Ÿ‡ฆ๐Ÿ‡บ๐Ÿ‡ฆdbcurren.bsky.social@bsky.brid.gy
2026-02-02

Q*Satoshi (@AiXsatoshi)

์ˆ™๋ จํ•œ ๊ธฐ์ˆ ์ž๊ฐ€ ๊ณ„์‚ฐ ์ž์›์„ ๋‹จ์ˆœํ•œ ์—๋„ˆ์ง€ ๋ณ€ํ™˜๊ธฐ๊ฐ€ ์•„๋‹ˆ๋ผ ์ง€๋Šฅ์„ ์ฆํญํ•˜๋Š” ์žฅ์น˜๋กœ ์ „ํ™˜ํ•œ๋‹ค๊ณ  ์ฃผ์žฅ. ๊ธฐ์ˆ  ํ˜์‹ ์ด ์—ฐ์‚ฐ ํšจ์œจ์„ ๋Œ์–ด์˜ฌ๋ฆฌ๋ฉด ์ƒ์„ฑ๋˜๋Š” ๊ฐ€์น˜๊ฐ€ ๊ธฐํ•˜๊ธ‰์ˆ˜์ ์œผ๋กœ ์ฆ๊ฐ€ํ•œ๋‹ค๋Š” ์ ์„ ๊ฐ•์กฐํ•˜๋Š” ๋‚ด์šฉ์ด๋‹ค.

x.com/AiXsatoshi/status/201810

#ai #compute #efficiency #scaling

2026-01-29

"Mแบกng computeprices.com ฤ‘ฦฐแปฃc ra mแบฏt ฤ‘แปƒ phรขn tรญch trung tรขm dแปฏ liแป‡u, huแบฅn luyแป‡n AI vร  thแป‹ trฦฐแปng thuรช cloud. Tuแบงn trฦฐแป›c, trang bแป• sung giรก mรด hรฌnh ngรดn ngแปฏ lแป›n (LLM) song song giรก GPU. ฤang nรขng cแบฅp theo dรตi giรก LLM & mแปŸ rแป™ng nguแป“n. Bแบกn muแป‘n tรญnh nฤƒng gรฌ thรชm? #Compute #AI #CloudComputing #TrungTรขmDแปฏLiแป‡u #MรดHรฌnhNgรดnNgแปฏLแป›n"

reddit.com/r/SideProject/comme

Startup Project (@startupprojectt)

Cerebras์˜ ์ตœ๊ณ ๊ฒฝ์˜์ง„์ด ์ œ์‹œํ•œ ํ†ต์ฐฐ: 70B(700์–ต) ํŒŒ๋ผ๋ฏธํ„ฐ๊ธ‰ ๋ชจ๋ธ์—์„œ ํ•œ ๋‹จ์–ด๋ฅผ ์ƒ์„ฑํ•˜๋ ค๋ฉด ์ „์ฒด ๊ธธ์ด ์˜ํ™” 100ํŽธ ๋ถ„๋Ÿ‰์— ํ•ด๋‹นํ•˜๋Š” ๋ฐ์ดํ„ฐ ์ด๋™์ด ํ•„์š”ํ•˜๋‹ค๋Š” ์ฃผ์žฅ. ์ด๋Š” ๋Œ€ํ˜• ์–ธ์–ด๋ชจ๋ธ(LLM) ์ถ”๋ก  ์‹œ ๋ˆˆ์— ๋„์ง€ ์•Š๋Š” ๋ง‰๋Œ€ํ•œ ๋ฐ์ดํ„ฐ ์ด๋™ ๋ฐ ๊ณ„์‚ฐ ๋น„์šฉ์„ ๊ฐ•์กฐํ•˜๋Š” ๋‚ด์šฉ์ž…๋‹ˆ๋‹ค.

x.com/startupprojectt/status/2

#cerebras #llm #inference #compute

investment data (@value_charts)

๊ธ€๋กœ๋ฒŒ AI ์ปดํ“จํŒ…(์—ฐ์‚ฐ) ์šฉ๋Ÿ‰์ด ์•ฝ 7๊ฐœ์›”๋งˆ๋‹ค ๋‘ ๋ฐฐ๋กœ ์ฆ๊ฐ€ํ•˜๊ณ  ์žˆ์œผ๋ฉฐ, ์ „๋ฐ˜์ ์ธ AI ์‚ฐ์—…์ด ๊ธฐํ•˜๊ธ‰์ˆ˜์  ์„ฑ์žฅ์„ธ๋ฅผ ๋ณด์ธ๋‹ค๋Š” ๊ด€์ธก์ž…๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ๊ธ‰์†ํ•œ ์ปดํ“จํŒ… ํ™•์žฅ์€ ๋ชจ๋ธ ํ›ˆ๋ จยท์ถ”๋ก  ๋Šฅ๋ ฅ ๋ฐ ์„œ๋น„์Šค ๋ฐฐํฌ ์—ญ๋Ÿ‰์„ ๋น ๋ฅด๊ฒŒ ํ™•๋Œ€ํ•ด ์—ฐ๊ตฌยท์ œํ’ˆํ™” ์†๋„์— ํฐ ์˜ํ–ฅ์„ ์ค„ ๊ฐ€๋Šฅ์„ฑ์ด ์žˆ์Šต๋‹ˆ๋‹ค.

x.com/value_charts/status/2014

#ai #compute #infrastructure #growth

Q*Satoshi (@AiXsatoshi)

์“ด ๊ธ€์˜ ์š”์ง€: ์Œ€๋†์‚ฌ ์‹œ๋Œ€์—๋Š” ํ† ์ง€๊ฐ€ ํ•ต์‹ฌ ์ž์‚ฐ์ด์—ˆ๋“ฏ์ด, ๋””์ง€ํ„ธ ์‹œ๋Œ€์—๋Š” ๊ณ„์‚ฐ(์ปดํ“จํŒ…) ์ž์›์ด ๋””์ง€ํ„ธ ๋Œ€์ง€์˜ ๊ธฐ๋ฐ˜์ด๋ผ๋Š” ๊ด€์ ์ด๋‹ค. ์—๋„ˆ์ง€๋งŒ์ด ์•„๋‹ˆ๋ผ ๊ณ„์‚ฐ ์ž์›์˜ ํ™•๋ณด์—๋„ ๊ตญ๊ฐ€ ์ „๋žต ์ˆ˜์ค€์˜ ์ž๊ธˆ ํˆฌ์ž๊ฐ€ ๊ณ„์†๋  ๊ฒƒ์ด๋ผ๋Š” ์ „๋ง์„ ์ œ์‹œํ•œ๋‹ค.

x.com/AiXsatoshi/status/201379

#compute #infrastructure #nationalpolicy #aiinfrastructure

OpenAI (@OpenAI)

AI์—์„œ ์—ฐ์‚ฐ(compute)์ด ๊ฐ€์žฅ ํฌ์†Œํ•œ ์ž์›์ด๋ฉฐ ์ˆ˜์š”๊ฐ€ ๊ณ„์† ์ฆ๊ฐ€ํ•œ๋‹ค๋Š” ๋…ผ์˜์ž…๋‹ˆ๋‹ค. OpenAI ํŒŸ์บ์ŠคํŠธ์—์„œ CFO ์‚ฌ๋ผ ํ”„๋ผ์ด์–ด์™€ Khosla Ventures ์ฐฝ์—…์ž Vinod Khosla๊ฐ€ ํ˜ธ์ŠคํŠธ Andrew Mayne์™€ ํ•จ๊ป˜ ์—ฐ์‚ฐ ์ˆ˜์š”, ์ธํ”„๋ผ ํ•œ๊ณ„, AI ํ˜œํƒ์„ ๋” ๋งŽ์€ ์‚ฌ๋žŒ์—๊ฒŒ ์ „๋‹ฌํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ๋…ผ์˜ํ•ฉ๋‹ˆ๋‹ค. ๊ฐœ๋ฐœยท์ธํ”„๋ผ ๊ด€์ ์—์„œ ์ฐธ๊ณ ํ•  ๋งŒํ•œ ๋Œ€ํ™”์ž…๋‹ˆ๋‹ค.

x.com/OpenAI/status/2013755708

#compute #openai #podcast #infrastructure

AIagent.at ๐Ÿค– AI Newsai@defcon.social
2026-01-19

#OpenAl says its #compute power grew from 0.2 GW in 2023 to 1.9 GW in 2025 and annualized #revenue grew from $2B in 2023 to $20B+ in 2025. OpenAIโ€™s #business model scales with the value delivered by intelligence, from #consumer subscriptions to #enterprise solutions and #commerce. openai.com/index/a-business-th #AIagent #AI #ML #NLP #LLM #GenAI

Thomas Cherryhomestschak@oldbytes.space
2026-01-18

This is my #Atari8Bit #NOS setup for writing my #Compute Articles with AtariWriter:

NOS from: TNFS://irata.online/Atari_8-bit/DOS/NOS.atr
AtariWriter in XEX form from atariwiki.org
AUTORUN.BAT in pic.

The NTRANS 2 tells #FujiNet to make LINUX ASCII text files. Windows users would use 1.
Save documents on N2: and they are readable on my PC.

TechRadar (@techradar)

Meta๊ฐ€ AI ์ธํ”„๋ผ ํ™•์žฅ์„ ๊ด€์žฅํ•  ์กฐ์ง 'Meta Compute'๋ฅผ ๋ฐœํ‘œํ–ˆ์Šต๋‹ˆ๋‹ค. ๋ฆฌ๋”์‹ญ์„ ํ†ตํ•ฉํ•ด AI ์ธํ”„๋ผ๋ฅผ ๊ด€๋ฆฌยทํ™•๋Œ€ํ•˜๊ณ  ํ–ฅํ›„ 10๋…„๊ฐ„ ์ˆ˜์‹ญ์—์„œ ์ˆ˜๋ฐฑ ๊ธฐ๊ฐ€์™€ํŠธ ๊ทœ๋ชจ์˜ ์ „๋ ฅ ๊ณ„ํš์„ ์„ธ์šฐ๋ฉฐ ๋Œ€๊ทœ๋ชจ ๋ฐ์ดํ„ฐ์„ผํ„ฐยท์ปดํ“จํŠธ ํ™•์ถฉ์„ ์ค€๋น„ํ•œ๋‹ค๋Š” ๋‚ด์šฉ์ž…๋‹ˆ๋‹ค. ํšŒ์‚ฌ ์ฐจ์›์˜ AI ์ธํ”„๋ผ ์ „๋žต๊ณผ ์—๋„ˆ์ง€ ์ˆ˜์š” ์ „๋ง์ด ํ•ต์‹ฌ์ž…๋‹ˆ๋‹ค.

x.com/techradar/status/2011611

#meta #ai #infrastructure #compute

Latent.Space (@latentspacepod)

์ตœ๊ทผ ๋ช‡ ๋…„ ๋™์•ˆ '์ง€๋Šฅ์˜ ๋น„์šฉ'์ด ๊ทน์ ์œผ๋กœ ํ•˜๋ฝํ•˜๊ณ  ์žˆ๋‹ค๋Š” ๊ด€์ฐฐ๋กœ, ์ง€๋Šฅ์„ ํ…Œ๋ผ ๋‹จ์œ„๋กœ ์ƒ๊ฐํ•  ๋•Œ ๋‹จ๊ฐ€๊ฐ€ ํฌ๊ฒŒ ๋–จ์–ด์กŒ์œผ๋ฉฐ ๊ทธ ๊ฒฐ๊ณผ GPT-4 ์ˆ˜์ค€์˜ ์ง€๋Šฅ์„ 100๋ฐฐ ์ด์ƒ ์ €๋ ดํ•˜๊ฒŒ ์–ป์„ ์ˆ˜ ์žˆ๋‹ค๋Š” ์ฃผ์žฅ์ž…๋‹ˆ๋‹ค. AI ๋ชจ๋ธ ์šด์šฉยท๊ฐœ๋ฐœ์˜ ๊ฒฝ์ œ์„ฑ์ด ํฌ๊ฒŒ ๊ฐœ์„ ๋˜๊ณ  ์žˆ์Œ์„ ์‹œ์‚ฌํ•˜๋Š” ์‚ฐ์—… ํŠธ๋ Œ๋“œ์ž…๋‹ˆ๋‹ค.

x.com/latentspacepod/status/20

#ai #gpt4 #cost #compute #scaling

TestingCatalog News (@testingcatalog)

Meta๊ฐ€ ์ƒˆ๋กœ์šด ์ „๋žต์  ์ด๋‹ˆ์…”ํ‹ฐ๋ธŒ 'Meta Compute'๋ฅผ ๋ฐœํ‘œ: ํ–ฅํ›„ 10๋…„ ๋™์•ˆ ์ธํ”„๋ผ๋ฅผ ์ˆ˜์‹ญ ๊ธฐ๊ฐ€์™€ํŠธ ๊ทœ๋ชจ๋กœ ํ™•์žฅํ•ด ์ปดํ“จํŠธ ์—ญ๋Ÿ‰์„ ํฌ๊ฒŒ ๋Š˜๋ฆฌ๊ฒ ๋‹ค๋Š” ๊ณ„ํš์„ ๊ณต๊ฐœํ•œ ์ „๋žต/์ธํ”„๋ผ ๊ด€๋ จ ๊ณต์ง€.

x.com/testingcatalog/status/20

#meta #infrastructure #compute #datacenter

H๐“ฎrmit เฝ€ (@hermitbuilder)

์ž‘์„ฑ์ž๊ฐ€ ์ปดํ“จํŠธ ์˜ˆ์ธก ๊ด€๋ จ ๊ธฐ์‚ฌ๋ฅผ ์†Œ๊ฐœํ•˜๋ฉฐ, 2026๋…„๊นŒ์ง€ ์—ฐํ‰๊ท  ์•ฝ 2-2.5๋ฐฐ ์„ฑ์žฅ์ด ์˜ˆ์ƒ๋˜์–ด ํ–ฅํ›„ 2๋…„ ๋‚ด ํ˜„์žฌ ๋Œ€๋น„ ์•ฝ 5-6๋ฐฐ์˜ ์ปดํ“จํŠธ ์šฉ๋Ÿ‰์ด ๊ฐ€๋Šฅํ•  ์ˆ˜ ์žˆ๋‹ค๋Š” ๋ถ„์„์„ ๊ณต์œ ํ–ˆ๋‹ค. ๊ด€๋ จ ๋งํฌ: ai-2027.com์˜ compute forecast ์—ฐ๊ตฌ.

x.com/hermitbuilder/status/201

#compute #forecast #ai #infrastructure

useideluseidel
2026-01-10

Yep! The 4 is a perfect replacement for the boards: running and quite some more

Client Info

Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst