Continuous Learning_Startup & Investment
2.44K subscribers
513 photos
5 videos
16 files
2.74K links
We journey together through the captivating realms of entrepreneurship, investment, life, and technology. This is my chronicle of exploration, where I capture and share the lessons that shape our world. Join us and let's never stop learning!
Download Telegram
----
Navigating "corporate speak" isn't easy.

Here's a helpful guide I put together:

"Let me check with my team" = No

"Possibly" = No

"On my roadmap" = Not happening

"This will be done in Q4" = This will be done in Q2 next year

"Disagree and commit" = I hate you

"Per my last email" = Try reading, for once in your life

"Challenging landscape" = We're going out of business, quickly

"Digital transformation" = We're going out of business, slowly

"Let's circle back" = We'll never speak of this again

"Take it offline" = We'll never speak of this again

"30,000 foot view" = I don't know what I'm saying

"Low hanging fruit" = Easy promotion

"Open up the kimono" = HR violation

"We use AI" = We don't use AI

"We use machine learning" = We don't use machine learning

"All hands on deck" = Let's actually try for once, please

๐Ÿ˜‚
๐Ÿ˜1
์•„์ดํฐ ์ถœ์‹œ ํ›„ ์ฒซ 4~5๋…„ ์ •๋„๋Š” ์•ฑ์„ ์ถœ์‹œํ•˜๋Š” ๊ฒƒ๋งŒ์œผ๋กœ๋„ ์œ ์ €์˜ wow๋ฅผ ์ด๋Œ์–ด ๋‚ผ ์ˆ˜ ์žˆ์—ˆ๋˜ ์‹œ์ ˆ์ด ์žˆ์—ˆ๋‹ค. ์ด๋™ํ•˜๋ฉฐ ์ธํ„ฐ๋„ท๋„ ํ•  ์ˆ˜ ์žˆ๊ณ  ํ„ฐ์น˜์Šคํฌ๋ฆฐ์„ ํ™œ์šฉํ•  ์ˆ˜ ์žˆ๋Š” ์Šค๋งˆํŠธํฐ์—์„œ ๊ตฌ๋™์ด ๋œ๋‹ค๋Š” ๊ฒƒ๋งŒ์œผ๋กœ๋„ wow ์˜€๋˜ ๊ฒƒ์ด๋‹ค.

๊ทธ ์ดํ›„ ๊ธฐ์กด ๋ ˆ๊ฐ€์‹œ ์‚ฐ์—…์˜ ํŒจ๋Ÿฌ๋‹ค์ž„์„ ๋ฐ”๊พผ ์„œ๋น„์Šค๋“ค์ด ๋“ฑ์žฅํ•˜์—ฌ wow ๋ฅผ ์„ ์‚ฌํ–ˆ๋‹ค. ์œ ํˆฌ๋ธŒ/๋„ทํ”Œ๋ฆญ์Šค๋Š” TV/๋ฏธ๋””์–ด ์‚ฐ์—…์„ ๋ฐ”๊ฟจ๊ณ , ํŽ˜๋ถ/๊ตฌ๊ธ€์€ ๋ฏธ๋””์–ด/ํฌํ„ธ์‚ฐ์—…์„ ๋ฐ”๊ฟ”๋†“์•˜์œผ๋ฉฐ, ์•„๋งˆ์กด์€ ์‡ผํ•‘/์„œ๋ฒ„์‹œ์Šคํ…œ์„, ํ…Œ์Šฌ๋ผ๋Š” ์ž๋™์ฐจ ์‚ฐ์—…์„, ์šฐ๋ฒ„๋Š” ํƒ์‹œ ์‚ฐ์—…์„, ์—์–ด๋น„์—”๋น„๋ฅผ ํ˜ธํ…” ์‚ฐ์—…์— ํฐ ์˜ํ–ฅ์„ ์คฌ๋‹ค.

์œ„์™€ ๊ฐ™์€ 2000๋…„ ์ดˆ๋ฐ˜ ์ดํ›„ ๊ฒฉ๋™์˜ 20๋…„ ์ •๋„๋ฅผ ๊ฒช์œผ๋ฉฐ, ํ˜„๋Œ€์‚ฌํšŒ์˜ ๋งŽ์€ ์‚ฌ๋žŒ๋“ค์€ ๋ชจ๋ฐ”์ผ ๋ฐ ์›น ์ค‘์‹ฌ์˜ ์„œ๋น„์Šค์— ๊ฝค ๋งŽ์ด ์ต์ˆ™ํ•ด์ง„ ๋“ฏ ํ•˜๋‹ค. ๊ทธ๋ž˜์„œ, ์ƒˆ๋กœ ์ถœ์‹œ๋˜๋Š” ์•ฑ์„ ๋ด๋„ ๊ทธ ๊ฐํฅ์ด ๊ณผ๊ฑฐ ๋Œ€๋น„ ๋ฌด๋ŽŒ์ง€๊ณ  ์žˆ์Œ์„ ๋А๋‚€๋‹ค.

์˜คํžˆ๋ ค, ์š”์ฆ˜์€ ๊ณผ๊ฑฐ์—๋Š” ๋งŒ์กฑํ•˜๋ฉฐ ์‚ฌ์šฉํ•˜๋˜ Big Tech ์‚ฌ์˜ ์„œ๋น„์Šค ์กฐ์ฐจ๋„ ์Šฌ์Šฌ '์ง€๊ฒจ์›Œ์ง„๋‹ค' ๋А๋ผ๋Š” ๋ถ„๋“ค์ด ์ฆ๊ฐ€ํ•˜๋Š” ๋“ฏํ•˜๋‹ค. ๋™์‹œ์— '๊ณผ๊ฑฐ๋ถ€ํ„ฐ ๋งŽ์ด ์ด์šฉํ•˜๋˜ ์„œ๋น„์Šค๋“ค์ด ์ตœ๊ทผ์—๋Š” ์˜ˆ์ „๋งŒ ๋ชปํ•ด์„œ ์‹ค๋ง๊ฐ์ด ํฌ๊ณ , ์–ด์ฉ” ์ˆ˜ ์—†์ด ์ด์šฉํ•˜๊ณ  ์žˆ๋‹ค'๊ณ  ๋งํ•˜๋Š” ๋ถ„๋“ค๋„ ๋งŽ์•„์ง€๋Š” ๋“ฏํ•˜๋‹ค. OpenAI ๊ฐ€ chatGPT ๋ฅผ ์ถœ์‹œํ•˜๋ฉฐ ์ƒˆ๋กœ์šด ํ™œ๋ ฅ์„ ๋ถˆ์–ด๋„ฃ๊ณ  ์žˆ์ง€๋งŒ, ๊ทธ ํŒŒ๊ธ‰๋ ฅ์ด ๊ณผ๊ฑฐ์˜ ์•„์ดํฐ ๋ชจ๋ฉ˜ํŠธ์ฒ˜๋Ÿผ ์—„์ฒญ๋‚˜์ง€ ์•Š์€ ๊ฒƒ๋„ ์‚ฌ์‹ค์ด๋‹ค. (chatGPT๋Š” ์†Œ๋น„์ž ๋ณด๋‹ค๋Š” ์Šคํƒ€ํŠธ์—…์— ํ™œ๋ ฅ์„ ๋ถˆ์–ด๋„ฃ๊ณ  ์žˆ๋Š” ๋“ฏํ•˜๋‹ค. ๊ทธ๋ž˜์„œ ๊ฑฑ์ •์ด๊ธฐ๋„ ํ•˜๋‹ค. ์†Œ๋น„์ž๊ฐ€ ๋ฐ˜์‘ํ•˜์ง€ ์•Š์œผ๋ฉด ๊ฒฐ๊ตญ imapct ๊ฐ€ ์ œํ•œ์ ์ผ ์ˆ˜๋ฐ–์— ์—†๊ธฐ ๋•Œ๋ฌธ์ด๋‹ค) ์š”์ฆ˜์€ ์œ ์ €๋ถ„๋“ค์ด ๋Œ€๋‹ค์ˆ˜ ์„œ๋น„์Šค์— ๊ณผ๊ฑฐ ๋Œ€๋น„ ๋œจ๋œจ๋ฏธ์ง€๊ทผํ•œ ๋ฐ˜์‘์„ ๋ณด์ด๋Š” ์‹œ๊ธฐ์ธ ๊ฒƒ ๊ฐ™๊ธฐ๋„ ํ•˜๋‹ค.

๋‹ค๋งŒ, ์œ ์ €๋“ค์€ ์—ฌ์ „ํžˆ ๋” ๋‚˜์€ ์‚ถ์„ ์‚ด๊ณ  ์‹ถ์–ดํ•˜๋Š” ์š•๊ตฌ๊ฐ€ ๊ฐ•ํ•˜๋‹ค. ๋” ์„ฑ์žฅํ•˜๊ณ  ์‹ถ์–ดํ•˜๊ณ , ๋” ํ–‰๋ณตํ•œ ์‚ถ์„ ์‚ด๊ณ  ์‹ถ์–ดํ•˜๊ณ , ๋” ์ฆ๊ฑฐ์šด ์‹œ๊ฐ„์„ ๋ณด๋‚ด๊ณ  ์‹ถ์–ด ํ•œ๋‹ค. ๋ณธ์ธ ์ธ์ƒ์— ๊ฐ•๋ ฌํ•˜๊ฒŒ ๋‹ค๊ฐ€์˜ฌ ์ˆ˜ ์žˆ๋Š” ์„œ๋น„์Šค๋ฅผ ๋งŒ๋‚˜, ํ•ด๋‹น ์„œ๋น„์Šค์™€ ํ•จ๊ป˜ ๋” ๋งŒ์กฑ์Šค๋Ÿฌ์šด ์‚ถ์„ ์‚ด๊ธฐ๋ฅผ ํฌ๋งํ•˜๋Š” ๋ถ„๋“ค์ด ๊ฝค ์žˆ๋‹ค.

๊ทธ๋ž˜์„œ, ๋ฏธ๋ž˜์—๋Š” ์–ด๋–ค ์„œ๋น„์Šค๊ฐ€ ๋งŽ์€ ์œ ์ €๋กœ๋ถ€ํ„ฐ wow๋ฅผ ์ด๋Œ์–ด ๋‚ผ ์ˆ˜ ์žˆ์„์ง€... ๊ถ๊ธˆํ•˜๋‹ค. ์œ ์ €์˜ ์„ฑ์žฅ๊ณผ ๋งŒ์กฑ์— ๋” ๋งŽ์ด ์ง‘์ฐฉํ•˜๋Š” ์„œ๋น„์Šค๊ฐ€ ๊ทธ wow ๋ฅผ ์ด๋Œ์–ด ๋‚ผ ์ˆ˜ ์žˆ์œผ๋ฆฌ๋ผ ๋ฏฟ์ง€๋งŒ, ๊ณผ๊ฑฐ ๋Œ€๋น„ ์œ ์ €์˜ wow ๋ฅผ ์ด๋Œ์–ด ๋‚ด๊ธฐ ์œ„ํ•œ ๋‚œ์ด๋„๊ฐ€ ๋†’์•„์ ธ์„œ ๊ณผ๊ฑฐ ๋งŒํผ '๊ธ‰์„ฑ์žฅํ•˜๋Š” ์Šคํƒ€ํŠธ์—…'์ด ๋งŽ์ด ๋‚˜์˜ฌ ๊ฒƒ ๊ฐ™์ง€๋Š” ์•Š๋‹ค. ๊ทธ๋ž˜๋„, ์ด๋Ÿฐ ๋‚œ์„ธ๋ฅผ ์ด๊ฒจ๋‚ด๋Š” ๊ฐ€์žฅ ์ข‹์€ ๋ฐฉ๋ฒ•์€ '๋” ํž˜๋“ค์–ด์ง€๋Š” ํ™˜๊ฒฝ'์— ๋ฐ˜์‘ํ•˜๊ธฐ ๋ณด๋‹ค๋Š” '์œ ์ €'์— ์ง‘์ค‘ํ•˜์—ฌ ํ•˜๋ฃจ ํ•˜๋ฃจ ๋” ์ข‹์€ ์„œ๋น„์Šค๋ฅผ ๋งŒ๋“ค์–ด ๋‚ด๋Š” ๊ฒƒ์ด๋ผ ์ƒ๊ฐํ•œ๋‹ค. ๋ฌผ๋ก  ๊ณผ๊ฑฐ์—๋Š” 1๋…„ ๋…ธ๋ ฅํ•˜๋ฉด ๋งŒ์กฑ์„ ์ด๋Œ์–ด ๋‚ผ ์ˆ˜ ์žˆ์—ˆ๋‹ค๋ฉด, ์ตœ๊ทผ์—๋Š” 3~5๋ฐฐ๋Š” ๋” ํ•ด์•ผํ•˜๋Š” ์–ด๋ ค์›€์ด ์กด์žฌํ•˜์ง€๋งŒ ๋ง์ด๋‹ค.

๊ทธ๋ž˜์„œ ์š”์ฆ˜ ๊ฐ™์€ ์‹œ๋Œ€๋Š” Why? ๊ฐ€ ๋” ์ค‘์š”ํ•ด์ง€๋Š” ๊ฒƒ ๊ฐ™๋‹ค. ์™œ ์‹œ์ž‘ํ–ˆ๋Š”์ง€? ๋ฌด์—‡์„ ์œ„ํ•ด ์„œ๋น„์Šค๋ฅผ ๋งŒ๋“ค์–ด ๋‚ด๊ณ  ์žˆ๋Š”์ง€?์— ๋Œ€ํ•œ ๋ช…ํ™•ํ•˜๊ณ  ์†”์งํ•œ ๋‹ต์„ ๊ฐ€์ง€๊ณ  ์žˆ๋Š”์ง€๊ฐ€ ๋” ์ค‘์š”ํ•ด์ง€๋Š” ์‹œ๋Œ€๊ฐ€ ์™”๋‹ค. ๊ทธ ์ด์œ ์— ๋Œ€ํ•œ ๋‹ต์„ ์ „์ฒด ํŒ€์ด ๊ณต์œ ํ•˜๊ณ  ์žˆ์–ด์•ผ, wow ๋ฅผ ๋งŒ๋“ค์–ด ๋‚ด๊ธฐ ์œ„ํ•œ ๋…ธ๋ ฅ/์‹œ๊ฐ„์ด ์ฆ๊ฐ€ํ•œ ์‹œ๋Œ€๋ฅผ ๋ฌต๋ฌตํžˆ ์ด๊ฒจ๋‚˜๊ฐˆ ์ˆ˜ ์žˆ๊ธฐ ๋•Œ๋ฌธ์ด๋‹ค.

์„œ๋น„์Šค๋ฅผ ์ด์–ด๋‚˜๊ฐ€๊ณ  ์žˆ๋Š” ๋ณธ์งˆ์  ์ด์œ ๋Š” ๋ฌด์—‡์ธ๊ฐ€? ์šฐ๋ฆฌ ์„œ๋น„์Šค๋Š” ๋ˆ„๊ตฌ๋ฅผ ์œ„ํ•ด ์กด์žฌํ•˜๋ฉฐ ์™œ ์กด์žฌํ•ด์•ผ ํ•˜๋Š”๊ฐ€? ๊นŠ์€ ๋ฐค, ๋ง๊ธ€์„ ๋” ์„ฑ์žฅ์‹œ์ผœ ๋‚˜๊ฐ€๊ธฐ ์œ„ํ•œ ๋ฐฉ๋ฒ•์„ ๊ณ ๋ฏผํ•˜๋‹ค๊ฐ€, ์›๋ก ์ ์ธ ์งˆ๋ฌธ์— ๋Œ€ํ•ด ๋‹ค์‹œ ํ•œ ๋ฒˆ ์ƒ๊ฐํ•ด๋ณธ๋‹ค.
โค3๐Ÿ‘1
โ€œMachine learning costs, talent and chip shortagesโ€ฆ any AI and machine learning company faces at least one of these challenges, and most face a few at a time,โ€ Pekhimenko told TechCrunch in an email interview. โ€œThe highest-end chips are commonly unavailable due to the large demand from enterprises and startups alike. This leads to companies sacrificing on the size of the model they can deploy or results in higher inference latencies for their deployed models.โ€

With spending on AI-focused chips expected to hit $53 billion this year and more than double in the next four years, according to Gartner, Pekhimenko felt the time was right to launch software that could make models run more efficiently on existing hardware.

โ€œTraining AI and machine learning models is increasingly expensive,โ€ Pekhimenko said. โ€œWith CentMLโ€™s optimization technology, weโ€™re able to reduce expenses up to 80% without compromising speed or accuracy.โ€

โ€œFor one of our customers, we optimized their Llama 2 model to work 3x faster by using Nvidia A10 GPU cards,โ€

CentML isnโ€™t the first to take a software-based approach to model optimization. It has competitors in MosaicML, which Databricks acquired in June for $1.3 billion, and OctoML, which landed an $85 million cash infusion in November 2021 for its machine learning acceleration platform.

โ€œThe CentML platform can run any model,โ€ Pekhimenko said. โ€œCentML produces optimized code for a variety of GPUs and reduces the memory needed to deploy models, and, as such, allows teams to deploy on smaller and cheaper GPUs.โ€
A new solution to the high-end chip shortage.

Read in @WSJ about how Together worked with large former crypto mining farms to repurpose their best GPUs and acquire new GPUs to train AI models โ€” all with a specialized training stack for a fraction of the price.

https://www.wsj.com/articles/crypto-miners-seek-a-new-life-in-ai-boom-after-an-implosion-in-mining-92a181fd
์–ผ๋งˆ์ „ ๊ตฌ๊ธ€ ๋ธŒ๋ ˆ์ธ, ๋”ฅ๋งˆ์ธ๋“œ์—์„œ ์ฝ”์–ด ๋”ฅ๋Ÿฌ๋‹ ํŒ€์— ์žˆ์—ˆ๋˜ ์นœ๊ตฌ์™€ ์ด์•ผ๊ธฐ ๋‚˜๋ˆ„๋ฉด์„œ ์ธ์ƒ์‹ถ์—ˆ๋˜ ๋ถ€๋ถ„๋“ค (์ด์„ธ์ข…๋‹˜)

์ง€๋‚œ๋ฒˆ ์„ธ์…˜์—์„œ ์ธ์ƒ๊นŠ์€ ๋ถ€๋ถ„์ž…๋‹ˆ๋‹ค.

quantization ๋“ฑ ๋ชจ๋ธ ์ตœ์ ํ™”๋‚˜ ๋ชจ๋ธ ์•„ํ‚คํ…์ฒ˜ ํšจ์œจํ™”๋กœ ๋ชจ๋ธ ํ•™์Šต/์ธํผ๋Ÿฐ์Šค ์ปดํ“จํŠธ(GPU) ๋น„์šฉ์ด ์ผ์‹œ์ ์œผ๋กœ ๊ฐ์†Œํ•  ์ˆ˜ ์žˆ์ง€๋งŒ, ์žฅ๊ธฐ์  ๊ด€์ ์—์„œ ์ปดํ“จํŠธ์˜ ์ ˆ๋Œ€์  ์ˆ˜์š”๋Š” ๋น ๋ฅด๊ฒŒ ์ฆ๊ฐ€ํ• ๊ฒƒ
- ๋”ฅ๋งˆ์ธ๋“œ์—์„œ ์•ŒํŒŒํด๋“œ๋ฅผ 3๋ช…์˜ ์—ฐ๊ตฌ์›์ด ํ•ด๋‚ผ ์ˆ˜ ์žˆ์—ˆ๋˜ ๊ฒƒ์€ ์ด๋“ค์ด ๋” ๋˜‘๋˜‘ํ•ด์„œ๊ฐ€ ์•„๋‹ˆ๋ผ 1์ธ๋‹น ํ™œ์šฉ ๊ฐ€๋Šฅํ•œ ์ปดํ“จํŠธ๊ฐ€ ๋‹ค๋ฅธ ๊ธฐ์—…, ์—ฐ๊ตฌ์†Œ ๋ณด๋‹ค ์••๋„์ ์œผ๋กœ ๋งŽ์•˜๊ธฐ ๋•Œ๋ฌธ
- ์ปดํ“จํŠธ๊ฐ€ ๋ณดํŽธํ™”๋˜์–ด ๊ฐ€๊ฒฉ์ด ์ €๋ ดํ•ด์งˆ ์ˆ˜๋ก ์•ŒํŒŒํด๋“œ ๊ธ‰์˜ ํ˜์‹ ์ด ๋ชจ๋“  ์‚ฐ์—…๊ณผ ์˜์—ญ์—์„œ ํŽผ์ณ์งˆ๊ฒƒ
- ๋‚˜์•„๊ฐ€ ์šฐ๋ฆฌ๋Š” ์ปดํ“จํŠธ์˜ ํ•œ๊ณ„๋กœ ๊ธฐ๋ณธ์ ์ธ ํ…์ŠคํŠธ ๋ฐ์ดํ„ฐ ํ”„๋กœ์„ธ์‹ฑ์— ๋ฐœ์ด ๋ฌถ์—ฌ ์žˆ์—ˆ๋Š”๋ฐ ์•ž์œผ๋กœ ์˜์ƒ ๋“ฑ ๋ณต์žก๋„ ๋†’๊ณ  ๋ฌด๊ฑฐ์šด ๋ฐ์ดํ„ฐ ์ฒ˜๋ฆฌ ์ˆ˜์š”๊ฐ€ ๊ธฐํ•˜๊ธ‰์ˆ˜์ ์œผ๋กœ ๋Š˜์–ด๋‚ ๊ฒƒ

์ด๋ฏธ ํ—ค์ง€ํŽ€๋“œ์—์„œ ์ฃผ๊ฐ€ ์˜ˆ์ธก์— ํŠนํ™”๋œ ๋ชจ๋ธ์„ ์ ๊ทน ํ™œ์šฉํ•ด ๋ง‰๋Œ€ํ•œ ์ˆ˜์ต ์ฐฝ์ถœ์ค‘
- ๋ชจ๋ธ ํ•™์Šต์— ์–ผ๋งˆ ๋“ค๊ณ , ๋ชจ๋ธ ๊ฒฝ์Ÿ์šฐ์œ„์˜ ์ง€์†์„ฑ๊ณผ ๊ธฐ๋Œ€ ๋งค์ถœ์„ ๊ณ ๋ คํ–ˆ์„ ๋•Œ ROI๊ฐ€ ์–ด๋–ป๊ฒŒ ๋ ์ง€๋ฅผ ๊ณ„์‚ฐํ•ด์„œ ๋™์‹œ ์—ฌ๋Ÿฌ๊ฐœ ๋ชจ๋ธ ํ•™์Šต์ค‘
- ๊ทธ์™ธ ๋ชจ๋“  ๊ธฐ์—…์—” ๋ ˆ๊ฑฐ์‹œ ์ฝ”๋“œ์™€ ๋ฐ์ดํ„ฐ๊ฐ€ ์žˆ๋Š”๋ฐ ๋ณด๊ด€๋œ ๋ฐฉ์‹๊ณผ ์ฝ”๋“œ์˜ ํ˜•ํƒœ ๋•Œ๋ฌธ์— ๋งค๋…„ ์–ด๋งˆํ•œ ๊ณ ์ •๋น„์šฉ์ด ๋ฐœ์ƒ. AI ๋ชจ๋ธ ๋„์›€์„ ๋ฐ›์•„ ์ฝ”๋“œ ์—…๋ฐ์ดํŠธ์™€ ๋งˆ์ด๊ทธ๋ ˆ์ด์…˜์„ ํ†ตํ•ด ํšŒ์‚ฌ ์ „๋ฐ˜์˜ ์ˆ˜์ต์„ฑ์„ ๊ฐœ์„ ํ•˜๋Š” ๊ฒฝ์šฐ๋„ ์ž์ฃผ ๋ณด์ž„

"์‚ฌ๋žŒ์„ ๋•Œ๋ ค๋ฐ•์•„" ์Šค์ผ€์ผ์„ ์ถ”๊ตฌํ•˜๋Š” ๋ธ”๋ฆฌ์ธ ์Šค์ผ€์ผ๋ง ๋ฐฉ๋ฒ•์€ AI ์‹œ๋Œ€์—์„œ ์œ ํšจํ•˜์ง€ ์•Š์„ ์ˆ˜๋„
- ๊ธฐ์—…์—์„œ ์ „ํ†ต์ ์œผ๋กœ ์ธ๊ฑด๋น„๊ฐ€ ๋ˆ์ด ๊ฐ€์žฅ ๋งŽ์ด ๋“ค์–ด๊ฐ”๋Š”๋ฐ ์ตœ๊ทผ์— ๊ตฌ๊ธ€์—์„  ์ปดํ“จํŠธ ๋น„์šฉ์ด ๊ฐœ๋ฐœ์ž ๋น„์šฉ์„ ์•ž์„œ
- ์˜คํ”ˆAI๋„ 400๋ช… ๋˜๋Š” ๊ธฐ์—…์ธ๋ฐ ์ปดํ“จํŠธ ๋น„์šฉ์ด ์ธ๋‹น ๊ฐœ๋ฐœ์ž ๋น„์šฉ์˜ 4๋ฐฐ
- ์•ž์œผ๋กœ๋Š” ์ž‘์ง€๋งŒ ํ”„๋กœ๋•ํŠธ๋ถ€ํ„ฐ ์„ธ์ผ์ฆˆ๊นŒ์ง€ ์ „์‚ฌ ์˜คํผ๋ ˆ์ด์…˜์— ์–ผ๋ผ์ธ๋œ ๋‹จ๋‹จํ•œ ํŒ€์ด ๊ฒฝ์Ÿ๋ ฅ ์žˆ์ง€ ์•Š์„์ง€

ํฌ๊ณ  ์ž‘์€ ์—ฌ๋Ÿฌ ์œ ํ˜•์˜ ์ œ๋„ˆ๋Ÿด๋ฆฌ์ŠคํŠธ์™€ ํŠนํ™”๋œ SOTA ๋ชจ๋ธ์ด ๊ณต์กดํ•˜๊ฒŒ๋ ๊ฒƒ
- ์œ ์ €๊ฐ€ ์›ํ•˜๋Š” ์š”์ฒญ์— ๋งž๋Š” ๋ชจ๋ธ์ด ์ž๋™ ์ถ”์ฒœ๋˜์–ด ํƒœ์Šคํฌ๊ฐ€ ์ฒ˜๋ฆฌ๋˜๋Š” Model of Experts ํ˜•ํƒœ
- ๋ชจ๋ธ์€ ํฌ๊ฒŒ ๋‘ ์ถ•์˜ ๊ต์ง‘ํ•ฉ์œผ๋กœ ์กด์žฌํ• ๊ฑด๋ฐ ํ•œ ์ถ•์€ intelligence (์˜ˆ: 80 IQ ~ 150 IQ), ๊ทธ๋ฆฌ๊ณ  ๋‹ค๋ฅธ ์ถ•์€ ๋ฒ„ํ‹ฐ์ปฌ (์˜ˆ: ์ฝ”๋”ฉ, ๋ฒ•๋ฅ , ์˜๋ฃŒ ๋“ฑ)
- ๋ชจ๋ธ ๊ฒ€์ƒ‰/์ถ”์ฒœ/์—ฐ๊ฒฐ์„ ๋•๋Š”orchestration ๋ ˆ์ด์–ด๊ฐ€ ํ•ต์‹ฌ ๊ธฐ์ˆ ๋กœ ๋ถ€์ƒํ• ๊ฒƒ

์ถœ์ฒ˜: ์ด์„ธ์ข…๋‹˜ ํŽ˜์ด์Šค๋ถ

https://www.facebook.com/Saejlee
๐Ÿ‘2
Continuous Learning_Startup & Investment
A new solution to the high-end chip shortage. Read in @WSJ about how Together worked with large former crypto mining farms to repurpose their best GPUs and acquire new GPUs to train AI models โ€” all with a specialized training stack for a fraction of the price.โ€ฆ
Today, Satoshi Spainโ€™s machines are handling AI workloads for startups, universities and individual developers in Europe.

โ€œYou can still make money from your mining rig,โ€ Satoshi Spain founder Alejandro Ibรกรฑez de Pedro said. โ€œItโ€™s mining 2.0.โ€

Around 20% of the freed-up chips can be repurposed to train AI models, said Vipul Ved Prakash, the CEO of alternative cloud provider Together.

Together is using former crypto miners for GPUs and other hardware to build virtual server farms. It has leased thousands of GPUs to help power Togetherโ€™s cloud service for AI developers. In May, it raised $20 million in seed funding.

Demi Guo, the founder of an AI applications startup called Mellis AI, couldnโ€™t get the data processing pricing or capacity she needed at the leading cloud providers so she is using repurposed hardware available through Togetherโ€™s cloud.

Saurabh Vij has built an AI business around chips formerly used for mining. His startup, Monster, has access to more than 30,000 chipsโ€”most of them once used for miningโ€”that its customers can lease.
Here's some AI โ€œmathโ€ from today's baseline:

- Chips improve by 2x next year
- Models improve by 2x next year
- Capex scales capacity by 10x next year

This means that whatever you see today will be ~40x better in a year and another ~20-40x better a year after that.

So we can expect an ~800-1,600x improvement by 2026.

What *wonโ€™t* be possible is probably the better question.
โค1
Continuous Learning_Startup & Investment
Here's some AI โ€œmathโ€ from today's baseline: - Chips improve by 2x next year - Models improve by 2x next year - Capex scales capacity by 10x next year This means that whatever you see today will be ~40x better in a year and another ~20-40x better a yearโ€ฆ
ChatGPT has some push back for you (see below). Its quite good!

The scenario you're describing is an incredibly optimistic one that assumes exponential growth across multiple dimensions of computing technology: chip performance, AI model capabilities, and capital expenditure for capacity scaling. In reality, while rapid advancements are certainly occurring, there are a number of physical, economic, and social constraints that could inhibit such a straightforward, exponential trajectory.

Hereโ€™s a breakdown of challenges and limitations in each of the areas you've mentioned:

1. Chip Performance: The idea that chip performance would continue to double every year is a significant acceleration of Moore's Law, which historically observed that the number of transistors on a microchip doubles about every two years. However, we're reaching physical limitations of silicon technology, with features approaching the scale of a few nanometers. Quantum tunneling, heat dissipation, and energy efficiency are all significant challenges. While new materials and paradigms (like quantum computing or 3D stacking) are in development, they also come with their own sets of challenges that might not allow for a simple 2x improvement year over year.

2. Model Improvements: AI models are indeed improving rapidly, but there's more to consider than raw performance. Larger models require more data, more power, and better algorithms to be effective. Improvements in algorithms could potentially hit a point of diminishing returns, and there are concerns about the environmental impact of training extremely large models. Additionally, large models can become increasingly difficult to manage, debug, and ensure they do not perpetuate biases or make erroneous decisions.

3. Capital Expenditure: A 10x scale in capital expenditure assumes not only that companies have the economic capacity and willingness to invest at such levels but also that the supply chains and raw materials necessary for such a scale-up are available. This also presumes that there will be sufficient market demand to justify this level of investment.

Now, letโ€™s address your main question: what won't be possible, even with such improvements? Here are some potential limitations:

- Physical Laws: Computing is still bound by the laws of physics. Even if computers become vastly more powerful, there will still be things that are simply impossible, like violating the speed of light or the second law of thermodynamics.

- Complexity and Emergent Phenomena: Many systems, like weather, human brains, or economies, are incredibly complex. Even with vastly improved models, the sheer number of variables and interactions can make prediction and understanding difficult.

- Social and Ethical Constraints: Just because something can be done technically does not mean it should be. Ethical considerations regarding surveillance, privacy, and the use of AI in decision-making will likely place limits on how technology is applied.

- Economic Considerations: Not every advancement will be economically feasible or provide a return on investment. Additionally, there could be resistance from industries that are disrupted by these advancements, potentially slowing adoption.

- Human Adaptability: The rate at which society can adapt to new technologies is limited. There are educational, cultural, and institutional barriers that often delay the implementation of even the most promising technologies.

In summary, while your scenario posits a world in which the capabilities of computing technology grow at an unprecedented rate, it's important to consider the multifaceted challenges that accompany such growth. The future is likely to be one of incredible technological advancement, but the path there will probably be more uneven and complex than the simple multiplication of current capabilities.
https://www.theinformation.com/articles/metas-free-ai-isnt-cheap-to-use-companies-say?rc=3jxaoo (์œ ๋ฃŒ๊ธฐ์‚ฌ์ž…๋‹ˆ๋‹ค)

์˜คํ”ˆ์†Œ์Šค ๋ชจ๋ธ์ด ์ƒ์—…๋ชจ๋ธ๋ณด๋‹ค ์ €๋ ดํ•˜์ง€ ์•Š์€ ๊ฒฝ์šฐ๊ฐ€ ํ›จ์”ฌ ๋งŽ๋‹ค๋Š” ๊ธฐ์‚ฌ์ž…๋‹ˆ๋‹ค. ๋‰ด์Šค์— ์ฐธ์กฐ๋œ ์•„๋ž˜์˜ ๊ทธ๋ž˜ํ”„ ๋ณด์‹œ๋ฉด ์‹ฌ์ง€์–ด 7B์งœ๋ฆฌ LLaMA ๋ชจ๋ธ ๊ตฌ๋™์ด GPT3.5 Turbo๋ณด๋‹ค ๋น„์‹ธ๋„ค์š”.

์ตœ๊ทผ LLM์˜ ๊ฒฝ๋Ÿ‰ํ™”/์ตœ์ ํ™” ์ถ”์„ธ๋ฅผ ๋ˆˆ์—ฌ๊ฒจ ๋ณด์‹œ๋Š” ๋ถ„๋“ค์ด๋ผ๋ฉด ์–ด์ฉŒ๋ฉด ๋‹น์—ฐํ•  ์ˆ˜๋„ ์žˆ๋Š” ๊ฒƒ์ด, ๊ต‰์žฅํžˆ ๋งŽ์€ traffic์ด ๋ชฐ๋ฆด ์ˆ˜๋ก (๋„๋งค ์žฅ์‚ฌ์ฒ˜๋Ÿผ) cost๋ฅผ ๋‚ฎ์ถœ ์ˆ˜ ์žˆ๋Š” ๋ฐฉ๋ฒ•๋“ค์ด ์กด์žฌํ•˜๋Š”๋ฐ, ์ปค์Šคํ„ฐ๋งˆ์ด์ง• ๋œ ์˜คํ”ˆ์†Œ์Šค ๋ชจ๋ธ์„ ๊ตฌ๋™ํ• ์ˆ˜๋ก ์„œ๋ฒ„ ๋น„์šฉ์ด ๋” ๋น„์‹ธ์ง€๊ณ , ์ฐจ๋ผ๋ฆฌ ๋ฐฑ๋ณธ ๋ชจ๋ธ์—์„œ tuning ์„œ๋น„์Šค๋ฅผ ๊ตฌ๋™ํ•˜๋Š” ๊ฒƒ์ด (์˜ˆ๋ฅผ ๋“ค๋ฉด LoRA weight๋“ค๋„ ๋™์ผ ๊ทœ๊ฒฉ์ด๋ฉด ํ•œ๊บผ๋ฒˆ์— ๋ฌถ์–ด๋ฒ„๋ ค์„œ ์„œ๋น„์Šค๋ฅผ ์ฒ˜๋ฆฌํ•œ๋‹ค๋˜๊ฐ€) ๊ฐ€๊ฒฉ์€ ๋” ์ €๋ ดํ•  ์ˆ˜ ๋ฐ–์— ์—†์Šต๋‹ˆ๋‹ค.

LLM์ด ํด๋ผ์šฐ๋“œ ์„œ๋น„์Šค์™€ ๋”๋”์šฑ ๋ฐ€์ ‘ํ•˜๊ฒŒ ์—ฐ๊ฒฐ๋  ์ˆ˜ ๋ฐ–์— ์—†๋‹ค๋ณด๋‹ˆ ์ด๋Ÿฌํ•œ ๊ฒฝํ–ฅ์ด ์ƒ๊ธธ ์ˆ˜ ๋ฐ–์— ์—†๋Š” ๊ฒƒ์ด๊ธฐ๋„ ํ•œ๋ฐ์š”, ๋„ค์ด๋ฒ„ํด๋ผ์šฐ๋“œ๊ฐ€ AI๋ฅผ ๋ฆฌ๋“œํ•˜๋Š” ๊ฒƒ์€ ์ด๋Ÿฌํ•œ ์ถ”์„ธ์— ์ ํ•ฉํ•œ ์„ ํƒ์ด๊ธฐ๋„ ํ–ˆ์Šต๋‹ˆ๋‹ค.

๋„ค์ด๋ฒ„ํด๋ผ์šฐ๋“œ๋Š” OpenAI๋‚˜ ๊ตฌ๊ธ€, ๋งˆ์ดํฌ๋กœ์†Œํ”„ํŠธ์™€ cost ๊ฒฝ์Ÿ์—์„œ ์ง€์ง€ ์•Š์„ ์—ฌ๋Ÿฌ๊ฐ€์ง€ ์ตœ์ ํ™”/๊ฒฝ๋Ÿ‰ํ™” ๊ธฐ์ˆ ๋“ค์„ ๊ตฌ๋™ํ•ด ๋†“์•˜๊ณ , ์ค‘์žฅ๊ธฐ์—์„œ๋Š” ์–ด๋А ๋น…ํ…Œํฌ๋ณด๋‹ค ๋›ฐ์–ด๋‚œ ์„œ๋น„์Šค ๊ฒฝ์Ÿ๋ ฅ์„ ๊ฐ–์ถœ AI ๋ฐ˜๋„์ฒด๋ฅผ ์ž˜ ์ง„ํ–‰ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
Continuous Learning_Startup & Investment
https://gpus.llm-utils.org/
์—ฌ๋Ÿฌ ์„œ๋น„์Šค๋ฅผ ๊ฝค ์ž˜ ๋น„๊ตํ•ด๋†จ๋„ค์š” ใ…Žใ…Ž

Which GPU cloud should I use?
#
If you need a huge number of A100s/H100s - talk to Oracle, FluidStack, Lambda Labs, maybe a few others. Capacity is very low though for large quantities, especially of H100s, based on a couple of cloud founders/execs Iโ€™ve talked with.
If you need a couple A100s or H100s: Runpod, perhaps Tensordock or Latitude.
If you need 1x H100: Runpod (Fluidstack and Lambda have been out of on-demand capacity for qhite a while).
If you need cheap 3090s, 4090s, or A6000s: Tensordock.
If you need Stable Diffusion inference only: Salad.
If you need a wide variety of GPUs: Runpod or Tensordock.
If you want to play around with templates / general hobbyist: Runpod.
The large clouds generally have worse pricing and more complicated setups than the above.
If youโ€™re tied to one of the big clouds (AWS, Azure, GCP), then you donโ€™t have a choice, so use that.
More info here, here, here, here, here and here.