The Strange Economics of LLM Inference-as-a-Service
From bycloud · · 23 min

The player loads from YouTube only when you click. It uses YouTube's privacy-enhanced mode.
If the video does not play, the owner may have turned off embedding. Watch it on YouTube instead.
Description
Try out Telnyx and use code BYCLOUD25 for $25 build credits! https://telnyx.com/sign-up?utm_source=bycloud&utm_medium=influencer&utm_campaign=202609_influencer_voiceai_glm53&utm_content=bycloud_yt_integration
In today's video, we are deep diving into the billion-dollar business of AI inference-as-a-Service, and find out how these companies make money serving open-source models by being faster, cheaper, and at massive scale.
Want to run large AI models on your private GPU servers? We can help you deploy and maximize inference performance. Book a free audit now at https://nimbuskernel.com/
my latest project: Intuitive AI Academy We just wrote a new piece on Linear Attention!! https://intuitiveai.academy/ limited time code "BACK2SCHOOL" for 50% off yearly plan
My Newsletter (weekly top research papers) https://mail.bycloud.ai/
My Patreon https://www.patreon.com/c/bycloud
Some sources if you wanna look more: [DeepSeek Serving] https://github.com/deepseek-ai/open-infra-index [Prefill-as-a-Service] https://arxiv.org/abs/2604.15039
Try out my new fav place to learn how to code https://scrimba.com/?via=bycloudAI
This video is supported by the kind Patrons & YouTube Members: 🙏Spam Maj, Alex, Chris LeDoux, DX Research Group, Poof N' Inu, Deagan, Robert Zawiasa, Ryszard Warzocha, Tobe2d, Louis Muk, Akkusativ, Kevin Tai, Mark Buckler, NO U, Tony Jimenez, Ângelo Fonseca, jiye, Anushka, Asad Dhamani, Binnie Yiu, Calvin Yan, Clayton Ford, Diego Silva, Etrotta, Gonzalo Fidalgo, Handenon, Hector, Jake Disco very, Michael Brenner, Nilly K, OlegWock, Daddy Wen, Shuhong Chen, Sid_Cipher, Stefan Lorenz, Sup, tantan assawade, Thipok Tham, Thomas Di Martino, Thomas Lin, Richárd Nagyfi, Paperboy, mika, Leo, Berhane-Meskel, Kadhai Pesalam, mayssam, Bill Mangrum, nyaa, Toru Mon, Lame Plane, Matej Macak, Len Mo, saylikhapekar, ZyanSheep, THEVIERAOS, Ricardo Raphael Corona-Moreno, superchordate
[Discord] https://discord.gg/NhJZGtH [Twitter] https://twitter.com/bycloudai [Patreon] https://www.patreon.com/bycloud [Business Inquiries] bycloud@smoothmedia.co [Other Inquiries] bycloudai@gmail.com [Profile & Banner Art] https://twitter.com/pygm7 [Video Editor] @aduckchicken2 Manim Animations created by https://x.com/miyashita_03 [Ko-fi] https://ko-fi.com/bycloudai [Bitcoin (BTC)] 3JFMJQVGXNA2HJE5V9qCwLiqy6wHY9Vhdx [Ethereum (ETH)] 0x3d784F55E0bE5f35c1566B2E014598C0f354f190 [Litecoin (LTC)] MGHnqALjyU2W6NuJSSW9fTWV4dcHfwHZd7 [Bitcoin Cash (BCH)] 1LkyGfzHxnSfqMF8tN7ZGDwUTyBB6vcii9 [Solana (SOL)] 6XyMCEdVhtxJQRjMKgUJaySL8cGoBPzzA2NPDMPfVkKN
0:00 Intro 2:07 Who serves LLMs? 3:55 What's Inference-as-a-Service? 4:55 The business model of IaaS 8:40 How to increase the margin? 9:36 Optimization: caching 11:28 Optimization: batching 12:56 Optimization: prefill-decode disaggregation 14:00 Optimization: custom kernel 19:01 Hardware optimization 20:36 conclusion