update
This commit is contained in:
parent
4eda5900c3
commit
51731ca419
@ -25,17 +25,25 @@ resource_packages:
|
||||
description: 适用于非常驻服务7b/14b-int4推理
|
||||
|
||||
- name: Ephemeral-RTX3090*1
|
||||
deploy_engines:
|
||||
- ascend_infer
|
||||
- vllm
|
||||
default_deploy_engine: vllm
|
||||
resources:
|
||||
gpu:
|
||||
type: gpu
|
||||
device_label: NVIDIA GeForce RTX 3090
|
||||
device_label: 3090
|
||||
gpu_key: nvidia.com/gpu
|
||||
gpu_value: 1
|
||||
cpu: 10
|
||||
cpu: 4
|
||||
memory: 20Gi
|
||||
description: 适用于非常驻服务7b/14b-int4推理
|
||||
- name: Ephemeral-RTX3090*2
|
||||
default_deploy_engine: vllm
|
||||
resources:
|
||||
gpu:
|
||||
type: gpu
|
||||
device_label: 3090
|
||||
gpu_key: nvidia.com/gpu
|
||||
gpu_value: 2
|
||||
cpu: 4
|
||||
memory: 20Gi
|
||||
description: 适用于非常驻服务7b/14b-int4推理
|
||||
- name: R200-8F*1
|
||||
|
||||
Loading…
Reference in New Issue
Block a user