Hacker News
- YaFSDP: a sharded data parallelism framework, faster for pre-training LLMs https://github.com/yandex/YaFSDP 16 comments
- Yet Another Way to Train Large Language Models https://github.com/yandex/YaFSDP 2 comments languagetechnology
- Recently, an improved version of the Fully Sharded Data Parallel (FSDP) library, YaFSDP, became publicly available. https://github.com/yandex/YaFSDP 2 comments learnmachinelearning
- Yandex open-sources YaFSDP, an LLM training tool that saves up to 20% of GPU resources. https://github.com/yandex/YaFSDP 7 comments technology
Linking pages
- Visualizing 6D Mesh Parallelism · main https://main-horse.github.io/posts/visualizing-6d/ 3 comments
- Yandex develops and open-sources YaFSDP — a tool for faster LLM training and optimized GPU consumption | by Mikhail Khrushchev | Yandex | Jun, 2024 | Medium https://medium.com/yandex/yafsdp-a-tool-for-faster-llm-training-and-optimized-gpu-utilization-is-no-632b7539f5b3 0 comments
- Top 9 Libraries to Accelerate LLM Building - by Avi Chawla https://www.blog.aiport.tech/p/top-9-libraries-to-accelerate-llm 0 comments
Linked pages
Would you like to stay up to date with Computer science? Checkout Computer science
Weekly.
Related searches:
Search whole site: site:github.com
Search title: GitHub - yandex/YaFSDP: YaFSDP: Yet another Fully Sharded Data Parallel
See how to search.