vllm/model_executor/models/llama4_eagle.py · 2bcf71b9c0305e2a3d645e7a5ced4460262b4a6c · OpenDAS / vllm_cscc · GitLab

Find file Blame History Permalink

[Spec Decode] Reduce TP communication for speculative decoding draft token generation (#34049) · 2bcf71b9
qizixi authored Feb 22, 2026
```
Signed-off-by: qizixi <qizixi@meta.com>
Co-authored-by: Lu Fang <30275821+houseroad@users.noreply.github.com>
```
2bcf71b9

llama4_eagle.py 9.4 KB