v1.0.4

78b09731 · chenzk · 670bcfcb · 78b09731 · 78b09731
Commit 78b09731 authored Nov 06, 2024 by chenzk
Hide whitespace changes
Inline Side-by-side

Showing with 2 additions and 2 deletions

README.md README.md +2 -2

whl/vllm-0.6.2+das.opt1.ac9aae1.dtk24042-cp310-cp310-linux_x86_64.whl ....2+das.opt1.ac9aae1.dtk24042-cp310-cp310-linux_x86_64.whl +0 -0

No files found.
--- a/README.md
+++ b/README.md
@@ -125,11 +125,11 @@ llamafactory-cli export examples/merge_lora/llama3_lora_sft.yaml
 llamafactory-cli chat examples/inference/llama3_lora_sft.yaml
 # 方法二：vllm 推理
 # 先安装新版vllm
-pip install whl/vllm-0.6.2+das.opt1.85def94.dtk24042-cp310-cp310-linux_x86_64.whl
+pip install whl/vllm-0.6.2+das.opt1.ac9aae1.dtk24042-cp310-cp310-linux_x86_64.whl
 pip install whl/flash_attn-2.6.1+das.opt2.08f8827.dtk24042-cp310-cp310-linux_x86_64.whl
-export LM_NN=0
 # 推理
 python infer_vllm.py # 后期可从光合开发者社区下载性能优化更好的vllm推理。
+# 若无法成功调用vllm，在终端输入命令：export LM_NN=0
 ```

 ## result

--- a/whl/vllm-0.6.2+das.opt1.85def94.dtk24042-cp310-cp310-linux_x86_64.whl
+++ b/whl/vllm-0.6.2+das.opt1.85def94.dtk24042-cp310-cp310-linux_x86_64.whl