vllm/lora/fully_sharded_layers.py · 1591c68fdea97a213d5564f687009c4fd1b44608 · OpenDAS / vllm_cscc · GitLab

Find file Blame History Permalink

[Kernel] Full Tensor Parallelism for LoRA Layers (#3524) · eefeb164
Austin Veselka authored Apr 27, 2024
```
Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
```
eefeb164

fully_sharded_layers.py 9.95 KB