How to make LLMs fast: KV Caching, Speculative Decoding, and Multi-Query Attention | Cursor Team

preview_player
Показать описание
See below for guest bio, links, and to give feedback, submit questions, contact Lex, etc.

*GUEST BIO:*
Aman Sanger, Arvid Lunnemark, Michael Truell, and Sualeh Asif are creators of Cursor, a popular code editor that specializes in AI-assisted programming.

*CONTACT LEX:*

*EPISODE LINKS:*

*SPONSORS:*
To support this podcast, check out our sponsors & get discounts:
*Encord:* AI tooling for annotation & data management.
*MasterClass:* Online classes from world-class experts.
*Shopify:* Sell stuff online.
*NetSuite:* Business management software.
*AG1:* All-in-one daily nutrition drinks.

*PODCAST LINKS:*

*SOCIAL LINKS:*
  1. Lex Clips
  2. alex friedman
  3. lex ai
  4. lex clips
  5. lex debate
  6. lex freedman
Рекомендации по теме
How to make 0:15:15

How to make LLMs fast: KV Caching, Speculative Decoding, and Multi-Query Attention | Cursor Team

All You Need 0:10:30

All You Need To Know About Running LLMs Locally

RUN LLMs on 0:01:59

RUN LLMs on CPU x4 the speed (No GPU Needed)

FREE Local LLMs 0:15:09

FREE Local LLMs on Apple Silicon | FAST!

Ollama: Run LLMs 0:06:06

Ollama: Run LLMs Locally On Your Computer (Fast and Easy)

How to Fine-Tune 0:10:41

How to Fine-Tune and Train LLMs With Your Own Data EASILY and FAST- GPT-LLM-Trainer

Mamba Might Just 0:14:06

Mamba Might Just Make LLMs 1000x Cheaper...

Using Clusters to 0:13:00

Using Clusters to Boost LLMs 🚀

Revolutionizing Transportation: Advanced 0:00:30

Revolutionizing Transportation: Advanced LLM Routing with RouteLLM

Run Local LLMs 0:15:05

Run Local LLMs on Hardware from $50 to $50,000 - We Test and Compare!

LLMs with 8GB 0:11:09

LLMs with 8GB / 16GB

Run LLMs locally 0:00:29

Run LLMs locally with LMStudio

I Ran Advanced 0:14:42

I Ran Advanced LLMs on the Raspberry Pi 5!

Run ALL Your 0:20:19

Run ALL Your AI Locally in Minutes (LLMs, RAG, and more)

Fast ReActions: Planning 0:15:16

Fast ReActions: Planning and Reasoning Quickly with LLMs

Speculative Decoding: When 0:12:46

Speculative Decoding: When Two LLMs are Faster than One

The EASIEST way 0:08:15

The EASIEST way to RUN Llama2 like LLMs on CPU!!!

How ChatGPT Works 0:07:54

How ChatGPT Works Technically | ChatGPT Architecture

How to Fine-Tune 0:15:15

How to Fine-Tune and Train LLMs With Your Own Data EASILY and FAST With AutoTrain

Run LLMs On 0:03:47

Run LLMs On Your Phone Locally - Easy & Fast Install

How might LLMs 0:22:43

How might LLMs store facts | DL7

Fine-tuning Large Language 0:28:18

Fine-tuning Large Language Models (LLMs) | w/ Example Code

What are Large 0:05:30

What are Large Language Models (LLMs)?

PyTorch in 100 0:02:43

PyTorch in 100 Seconds

Комментарии
Автор

See below for guest bio, links, and to give feedback, submit questions, contact Lex, etc.

*GUEST BIO:*
Aman Sanger, Arvid Lunnemark, Michael Truell, and Sualeh Asif are creators of Cursor, a popular code editor that specializes in AI-assisted programming.

*CONTACT LEX:*

*EPISODE LINKS:*

*SPONSORS:*
To support this podcast, check out our sponsors & get discounts:
*Encord:* AI tooling for annotation & data management.
*MasterClass:* Online classes from world-class experts.
*Shopify:* Sell stuff online.
*NetSuite:* Business management software.
*AG1:* All-in-one daily nutrition drinks.

*PODCAST LINKS:*

*SOCIAL LINKS:*

LexClips
Автор

Imagine human behaviour was relative to the time zones they lived in and you could guarantee 5 Ti gpus running in each household every morning to heat water for their showers like Santa flying around the world doing computations while heating on demand 😂

ValidatingUsername
Автор

I really like the channel but I had to unsubscribe because my feed is absolutely flooded with short clips from this channel making it annoying as fuck to find anything else

saintsplenetic