ChengShuting comments

Repositories
Issues
Comments

Results 2 comments of


                                            ChengShuting

How to fix incomplete answers?

I also encountered this problem. Does anyone has any idea?

How to solve the problem of errors when loading qwen1.5-7B (using two GPUs) and llama3-8B (using two GPUs) models simultaneously using tritonserver?

> Is it necessary to launch these two models together? Could you use > > ```shell > tritonserver --model-control-mode=explicit --modelepository=/data/multi_model_repo/ --load-model=Qwen1.5-7B-Chat > tritonserver --model-control-mode=explicit --modelepository=/data/multi_model_repo/ --load-model=Llama3-8B-Chinese-Chat > ``` > >...