Publications

3 results for Jamie Yang

vllm-triton-backend: How to get state-of-the-art performance on NVIDIA and AMD with just triton
- - Burkhard Ringlein
  - Thomas Parnell
  - et al.
- 2025
- PyTorch Conference 2025
The Anatomy of a Triton Attention Backend
- - Burkhard Ringlein
  - Jan van Lunteren
  - et al.
- 2025
- Triton Developer Conference 2025
Triton in Action: Real-World Optimizations for Mamba2 and vLLM
- - Jamie Yang
  - Sara Kokkila Schumacher
  - et al.
- 2025
- Triton Developer Conference 2025