kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
32dfdfba30
TensorRT-LLMs/docs/source/advanced/images
History
kanghui0204 6f3922f318
feat: Low Precision Allreduce for PCIe based GPU (#4344) 
This PR adds a customized allreduce to TensorRT-LLM. The new allreduce is used for communication on PCIe-based GPUs via low-precision quantization, which can accelerate the PCIe allreduce process.

Signed-off-by: Hui Kang <hkang@nvidia.com>
Co-authored-by: Hui Kang <hkang@nvidia.com>
2025-05-20 06:53:46 +08:00
..
8x_l20_L40S_node_architecture.png feat: Low Precision Allreduce for PCIe based GPU (#4344) 2025-05-20 06:53:46 +08:00
disaggregated-service_usage.png Update TensorRT-LLM (#2755) 2025-02-11 03:01:00 +00:00