February 2026
Intermediate to advanced
406 pages
13h 25m
Italian
Content preview from IA generativa su Kubernetes (Italian Edition)
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.
Capitolo 2. Datidel modello
Questo lavoro è stato tradotto utilizzando l'AI. Siamo lieti di ricevere il tuo feedback e i tuoi commenti: translation-feedback@oreilly.com
Una delle sfide più grandi quando si gestiscono i modelli di linguaggio grande (LLMs) su Kubernetes è gestire la quantità enorme di dati del modello. I modelli LLM possono avere dimensioni che vanno da pochi gigabyte a quasi un terabyte, e portare questi dati in modo efficiente in un cluster dove i runtime possono accedervi richiede un'attenta valutazione.
La parte principale di questi modelli è costituita dai parametri del modello e può essere estremamente grande. La tabella 2-1 elenca il numero di parametri e le dimensioni di alcuni dei modelli più importanti disponibili che è possibile eseguire autonomamente. Ce ne sono molti altri, ma da questa selezione è già possibile notare un'ampia gamma di variazioni. Si va da modelli di grandi dimensioni, probabilmente poco pratici per un utilizzo on-demand, a modelli più leggeri che possono essere eseguiti sul proprio cluster e facilmente scaricati quando necessario.
| Nome | Fornitore | Parametri | Dimensione |
|---|---|---|---|
Llama 4 Maverick |
Meta |
400 miliardi (MoE, 17 miliardi attivi) |
~800 GB |
DeepSeek-V3 |
DeepSeek |
671 miliardi (MoE, 37 miliardi attivi) |
~700 GB |
Llama 3.1 405 miliardi |
Meta |
405 miliardi |
~750 GB |
Qwen3-235B |
Alibaba |
235 miliardi (MoE, 22 miliardi attivi) |
~118 GB |
Mixtral 8x22 miliardi |
Mistral |
141 miliardi (MoE, 39 ... |
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access
More than 5,000 organizations count on O’Reilly
O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.Julian F.
I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.Addison B.
I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.Amir M.
I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.