bjwswang

support multimodal models in model service

2

Before we enable multimodal in kubeagi,we should be able to manage and deploy multi-modal models. For multimodal models,we can focus on test-image. - Qwen-VL: https://github.com/QwenLM/Qwen-VL/blob/master/README_CN.md - CogVLM https://github.com/THUDM/CogVLM

Knowledgeretriever can add a new field `KnowledgebaseRef` _Originally posted by @bjwswang in https://github.com/kubeagi/arcadia/issues/475#issuecomment-1888614961_

Sometimes worker is pending due to resouce request not enough

When user creates multiple konwledgebases , they might use them in one application.

we should return the knowledgebase ready status to the user, so the user can know whether it's fine to chat with docs. _Originally posted by @nkwangleiGIT in https://github.com/kubeagi/arcadia/pull/887#discussion_r1530528514_