The LAS service provides VeDaft images that can be used directly in LAS development environments such as development machines. These images include Daft, as well as dependencies and runtime environments related to multimodal operators. The VeDaft image list is as follows.
Image name/ID | Image description | Framework | Machine type | Image size | Image repository address |
|---|---|---|---|---|---|
ve-daft | ve-Daft is a customized version of Daft launched by the Volcano Engine LAS team. It includes Daft, as well as dependencies and runtime environments related to multimodal operators, such as Cuda, Conda, Ray, and many others. | Daft | CPU、GPU | 16G | las-ai-cn-beijing.cr.volces.com/las/ve-daft:cu12.4.1-py3.11-0.5.8.post100 |
tip
For more official images provided by LAS, see Image management.
The EMR service provides a list of images used by the Daft service, including Daft, as well as dependencies and runtime environments related to multimodal operators. These images can be used directly on Volcano Engine ECS, making it convenient for you to perform further development in your own Volcano Engine ECS development environment based on these images.
Region (Chinese name) | Region (English name) | Image repository name | Image size | Notes on dependencies |
|---|---|---|---|---|
North China | cn-beijing | emr-vke-public-cn-beijing.cr.volces.com/emr/ve-daft:0.7.2.post6-py3.11-ubuntu24.04 | 11.87 GB |
|
North China | cn-beijing | emr-vke-public-cn-beijing.cr.volces.com/emr/ve-daft:0.7.2.post6-cu12.9.1-py3.11-ubuntu24.04 | 6.22 GB |
|
tip