Skip to main content

Arcadia Configuration File

We have some system level configuration for Arcadia, and we can update it based on our requirements.

Here is a sample configuration file

    # RelationalDatasource specifies the built-in datasource(common:postgres) for Arcadia to host relational data
relationalDatasource:
apiGroup: arcadia.kubeagi.k8s.com.cn/v1alpha1
kind: Datasource
name: 'arcadia-postgresql'
namespace: 'kubeagi-system'
# System datasource that will hold all models and datasets
systemDatasource:
apiGroup: arcadia.kubeagi.k8s.com.cn/v1alpha1
kind: Datasource
name: 'arcadia-minio'
namespace: 'kubeagi-system'
# Fastchat gateway to serve all local model services
gateway:
apiServer: 'http://arcadia-fastchat.kubeagi-system.svc.cluster.local:8000/v1'
externalApiServer: 'http://fastchat.172.40.20.125.nip.io/v1'
controller: 'http://arcadia-fastchat.kubeagi-system.svc.cluster.local:21001'
# Vectorstore for embedding
vectorStore:
apiGroup: arcadia.kubeagi.k8s.com.cn/v1alpha1
kind: VectorStore
name: 'arcadia-pgvector-vectorstore'
namespace: 'kubeagi-system'
# Resource pool managed by Ray cluster
rayClusters:
- name: 3090-2-GPUs
headAddress: raycluster-kuberay-head-svc.kuberay-system.svc:6379
pythonVersion: 3.9.18
# Streamlit configuration for each namespace if enabled, disabled by default
# MUST enable these configuration if enable streamlit for namespace, refer to Quick Start -> Run LLM application using Streamlit
streamlit:
image: kubeagi/streamlit:v1.29.0
ingressClassName: portal-ingress
host: portal.172.40.20.125.nip.io
contextPath: /chats
# Embedder specifies the default embedder for Arcadia to generate embeddings
embedder:
apiGroup: arcadia.kubeagi.k8s.com.cn/v1alpha1
kind: Embedder
name: 'bge-large-zh'
namespace: 'ainet'
# the default endpoint of rerank
rerankDefaultEndpoint: "http://172.31.52.30/api/v1/rerank"