New RunLLM response generated: >>>*# Question* Hi there, I'm trying to figure out a good developer flow for our data scientists where we don't need to specify packages in image spec and other pre-existing places such as `requirements.txt`/poetry etc. What is the recommended way of solving this? Ideally I don't want to have to pass

container_image=...

to every task. Calling`pyflyte` run/register with

--image

and an image we created using an appropriate docker file works just fine. But what I was hoping to be able to do was to add something like:

images:
  default: localhost:30000/myimage:latest

to the config (on the client or server) and then just have things work? # Answer To set a default image for Flyte tasks without specifying

container_image

for each task, you can configure the image in the

$HOME/.flyte/config.yaml

file. According to Flyte documentation, you can include an "images" section in your configuration file to specify default images. This allows you to avoid passing the

--image

flag every time you run or register a workflow. Example configuration:

admin:
  endpoint: localhost:30080
  authType: Pkce
  insecure: true
images:
  default: localhost:30000/myimage:latest

Would you like more detail on how to implement this configuration or any other aspect of Flyte's image handling? Sources: - Flyte documentation Slack Link https://.slack.com/archives/CP2HDHKE1/p1729867407.888849 👍 0 button 👎 0 button 💡 Hint: Mention @runllm in the thread for followups.