« KVarN » : différence entre les versions
De Le Wiki de Lug
Autres actions
| Ligne 5 : | Ligne 5 : | ||
# git clone <nowiki>https://</nowiki>github.com/huawei-csl/KVarN.git | # git clone <nowiki>https://</nowiki>github.com/huawei-csl/KVarN.git | ||
# cd KVarN | # cd KVarN | ||
# VLLM_USE_PRECOMPILED=1 pip install -e . | |||
# VLLM_USE_PRECOMPILED=1 | |||
Vérification : | Vérification : | ||
# python -c "import vllm, vllm._C; print(vllm.__file__); print(vllm._C.__file__)" | # python -c "import vllm, vllm._C; print(vllm.__file__); print(vllm._C.__file__)" | ||
Version du 26 juillet 2026 à 09:55
Installation
# cd /opt/ # git clone https://github.com/huawei-csl/KVarN.git # cd KVarN # VLLM_USE_PRECOMPILED=1 pip install -e .
Vérification :
# python -c "import vllm, vllm._C; print(vllm.__file__); print(vllm._C.__file__)"
Tests en vrac
# hf download shawnw3i/Qwen3.6-27B-AWQ-MTP --local-dir /opt/models/Qwen3.6-27B-AWQ-MTP