vllm/model_executor/models/llama.py · 1d094fd7c0d72e7eaf3a2b65fff9f7de8d812a49 · OpenDAS / vllm_cscc · GitLab

Find file Blame History Permalink

[Distributed][PP] only create embedding & lm head when necessary (#6455) · 1d094fd7
Wushi Dong authored Jul 16, 2024
```
original title: [Distributed][Model] Rank-based Component Creation for Pipeline Parallelism Memory Optimization
```
1d094fd7

llama.py 20 KB