Stas Bekman

stas

https://stasosphere.com/machine-learning/

AI & ML interests

Toolmaker. Software creator, optimizer and harmonizer. Makes things work and fly at Contextual.AI Training LLM/RAG/Generative AI/Machine Learning/Scalability

Articles

From DeepSpeed to FSDP and Back Again with Hugging Face Accelerate

Jun 13

• 44

Introducing IDEFICS: An Open Reproduction of State-of-the-art Visual Language Model

Aug 22, 2023

• 26

Organizations

stas's activity

upvoted an article 2 months ago

Article

Mixture of Experts Explained

Dec 11, 2023

• 176

upvoted an article 5 months ago

Article

From DeepSpeed to FSDP and Back Again with Hugging Face Accelerate

Jun 13

• 44

upvoted an article 7 months ago

Article

Introducing Idefics2: A Powerful 8B Vision-Language Model for the community

Apr 15

• 164

Stas Bekman

AI & ML interests

Articles

From DeepSpeed to FSDP and Back Again with Hugging Face Accelerate

Introducing IDEFICS: An Open Reproduction of State-of-the-art Visual Language Model

Incredibly Fast BLOOM Inference with DeepSpeed and Accelerate

The Technology Behind BLOOM Training

Fit More and Train Faster With ZeRO via DeepSpeed and FairScale

Porting fairseq wmt19 translation system to transformers

Organizations

stas's activity

Mixture of Experts Explained

From DeepSpeed to FSDP and Back Again with Hugging Face Accelerate

Introducing Idefics2: A Powerful 8B Vision-Language Model for the community