Hi All! I want to be the first one to say hi here! 👋 I am excited about the announcement. I would like to learn more about Yatai and hope to contribute where ever I am qualified for 🙂 BentoML is awesome. We use it for work and non-profit!

hello~

I tried to use yatai in k8s

Is there any way to use node affinity in yatai?

because I want to select gpu type to be deployed

can i change admin account for yatai?

What’s the long-term destiny of the Yatai project? Is maintenance of it going to be discontinued as BentoML focuses on BentoCloud? Or will it continue to be “mini-BentoCloud” without the full features? Also, can anyone who has tried running Yatai in production comment on how the experience has been? (do bugs get support? Are there bugs? Do you find it a good experience?) I’m not a kubernetes guy so I don’t have a personal stake in this, but I think this is important to know for when I’m trying to pitch BentoML to wherever I work.

hello all, i was wondering if this issue has any plans on being resolved: https://github.com/bentoml/Yatai/issues/509. We are experiencing the same memory leak on 1.1.13

Hello, I am trying to run a demo of Yatai to my organization on our GPU infrastructure. So far I have setup Yatai successfully. When I build and push a

summarization

bento then try to run the github example

bentoDeployment

and

BentoRequest

objects get created but nothing else happens. When I look at the events on the bentodeployment, I see :

I also can see this deployment in the UI, but when I get deployments.resources from the cli I dont see anything. What am I missing ?:

Do I need to build the bento with the

--version

flag set to an older version of bentoml ? if so which one ?

thanks

When I do it through the GUI instead of following the tutorial I see this :