Latent Space: The AI Engineer Podcast — FlashAttention 2: making Transformers 800% faster w/o approximation - with Tri Dao of Together AI

SCOUT· 65%54m·2023-07-26·Technology

FlashAttention 2: making Transformers 800% faster w/o approximation - with Tri Dao of Together AI

Latent Space: The AI Engineer Podcast

FlashAttention was first published by Tri Dao in May 2022 and it had a deep impact in the large language models space. Most open models you’ve heard of (RedPajama, MPT, LLaMA, Falcon, etc) all leverage it for faster inference. Tri came on the podcast to chat about FlashAttention, the newly released FlashAttention-2,…

Listen now

Buzzmeter · cross-platform

65%

SCOUT

Scout bees are checking it out — divided opinions, the swarm hasn't committed.

Hive Vote · Skep listeners

New on SkepAwaiting the hive's first votes

Buzzmeter says it's buzzing across platforms — the Hive Vote says whether the people who actually listened liked it.

Rate this episode

Add your vote to the Hive — listeners rate every episode after they finish.

Recent reviews

No reviews yet — be the first.

FlashAttention 2: making Transformers 800% faster w/o approximation - with Tri Dao of Together AI

Rate this episode

More from Latent Space: The AI Engineer Podcast

Why AI Infrastructure must evolve for Agent Experience — Akshat Bubna, Modal CTO

🔬Doing Vibe Physics — Alex Lupsasca, OpenAI

🔬 Training Transformers to solve 95% failure rate of Cancer Trials — Ron Alfa & Daniel Bear, Noetik