Memuat…

Course Forum Asisten Sumber Cara belajar RAG Peta modul

Masuk Mulai →

socrates

Semua tentang AI, satu jalur belajar. Gratis, bersumber terbuka, dan setiap jawaban punya rujukan.

Navigasi

Course
Forum
Asisten
Sumber
Cara belajar
RAG
Peta modul

Dibangun dengan

Next.js · Tailwind ·
Lenis + GSAP · SQLite + BM25
(RAG hybrid: cosine + BM25, RRF)

© 2026 socratesgratis · terbuka · sumber selalu dicantumkan

I2 · Inference Engines: vLLM, PagedAttention & Continuous Batching — socrates · socrates

LLM Infrastructure/Lintasan Inti/I2

I2

Inference Engines: vLLM, PagedAttention & Continuous Batching

Arsitektur vLLM, PagedAttention, continuous batching, chunked prefill, disaggregated serving

Durasi

1.5 mgg

Lesson

5

Project

1

Mode

3 jenis

Lesson

0/5 selesai

01
Anatomi vLLM: Mesin Inferensi LLM Skala Produksi
02
PagedAttention: Manajemen Memori KV Cache Tanpa Fragmentasi
03
Continuous Batching (Iteration-Level) vs Static Batching
04
Disaggregated Serving: Memisahkan Prefill & Decode Node
05
Project: Benchmark vLLM Server & Concurrency Stress Test

← I1GPU Kernels: Dari CUDA ke Triton & FlashAttention I3 →Distributed Serving & Parallelism Strategies