×

Home Discussions Write at Opengenus IQ

×

☰

DSA Cheatsheet
HOME
Jobs
About
One Liner
RANDOM
Track your progress
Deep Learning (FREE)
Join our Internship 🎓
AI Engineering
Scholarship
Campus Ambassador

Search anything:

Interesting posts

Top Competitive Programmers
Unsolved Problems in Algorithms
Top patent holders in India
How to get Developer Job in Japan?
[INTERNSHIP]
STORY: Most Profitable Software Patents
How to earn by filing Patents?
Richest Programmers in the World
STORY: Multiplication from 1950 to 2022
Position of India at ICPC World Finals (1999 to 2021)
Most Dangerous Line of Code 💀
Age of All Programming Languages
How to earn money online as a Programmer?
STORY: Kolmogorov N^2 Conjecture Disproved
STORY: man who refused $1M for his discovery
STORY: Man behind VIM
STORY: Galactic algorithm
STORY: Inventor of Linked List
Practice Interview Questions
List of 50+ Binary Tree Problems
List of 100+ Dynamic Programming Problems
List of 50+ Array Problems
11 Greedy Algorithm Problems [MUST]
List of 50+ Linked List Problems
100+ Graph Algorithms and Techniques

AI course essential tutorial

5D Parallelism
Memory components in LLM
P/D disaggregation
Continuous Batching
Decode-Maximal Batching
Chunked Prefill
Layerwise timing of LLM
LLM Inference
Lifecycle Aware Performance
System-Level Parallelism & Scaling [AI workload]
INT8 Quantization
[Calculate] Communication and compute overhead
[Calculate] LLM memory calculations
Global AI Race

LLM Inference

AI course essential

This post is for subscribers only

Subscribe now

Already have an account? Sign in

This chapter is part of the AI Engineering Course

You're reading a preview. The full chapter, worked benchmarks, and 1:1 guidance are part of the OpenGenus AI Engineering Course.

See the course → View the free roadmap

Subscribe to OpenGenus IQ: Learn Algorithms, DL, System Design

Get the latest posts delivered right to your inbox

OpenGenus

AI course essential

5D Parallelism
Memory components in LLM
P/D disaggregation

See all 14 posts →

AI course essential

Layerwise timing of LLM

[CHEATSHEET] Understand the overhead of LLM Inference.

OpenGenus

AI course essential

Lifecycle Aware Performance

In safety-critical, long-lifecycle systems = obsolescence is the real bottleneck. Adaptive architectures (FPGAs, SoCs, stable CPUs) = keep innovation alive after deployment.

OpenGenus

OpenGenus IQ: Learn Algorithms, DL, System Design

—

LLM Inference