NNTile is a framework for training large neural networks. It relies on a task-based parallel programming paradigm, which distributes computations across all avialable hardware resources dynamically and transmits data asynchronously. For this purpose NNTile utilizes StarPU library.
Pinned Loading
Repositories
Showing 5 of 5 repositories
- starpu Public Forked from starpu-runtime/starpu
This is a mirror of https://gitlab.inria.fr/starpu/starpu where our development happens, but contributions are welcome here too!
nntile/starpu’s past year of commit activity - litgpt Public Forked from Lightning-AI/litgpt
20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.
nntile/litgpt’s past year of commit activity
Top languages
Loading…
Most used topics
Loading…