ChatGPT agent¶

This guide provides an overview of how to set up the ChatGPT agent in your Flyte deployment. Please note that you have to set up the OpenAI API key in the agent server to to run ChatGPT tasks.

Specify agent configuration¶

Edit the relevant YAML file to specify the agent.

kubectl edit configmap flyte-sandbox-config -n flyte

tasks:
  task-plugins:
    enabled-plugins:
      - container
      - sidecar
      - k8s-array
      - agent-service
    default-for-task-types:
      - container: container
      - container_array: k8s-array
      - chatgpt: agent-service

plugins:
  agent-service:
    supportedTaskTypes:
    - chatgpt
    # Configuring the timeout is optional.
    # Tasks like using ChatGPT with a large model might require a longer time,
    # so we have the option to adjust the timeout setting here.
    defaultAgent:
      timeouts:
        ExecuteTaskSync: 10s

Create a file named values-override.yaml and add the following configuration to it:

configmap:
  enabled_plugins:
    # -- Tasks specific configuration [structure](https://pkg.go.dev/github.com/flyteorg/flytepropeller/pkg/controller/nodes/task/config#GetConfig)
    tasks:
      # -- Plugins configuration, [structure](https://pkg.go.dev/github.com/flyteorg/flytepropeller/pkg/controller/nodes/task/config#TaskPluginConfig)
      task-plugins:
        # -- [Enabled Plugins](https://pkg.go.dev/github.com/flyteorg/flyteplugins/go/tasks/config#Config). Enable sagemaker*, athena if you install the backend
        enabled-plugins:
          - container
          - sidecar
          - k8s-array
          - agent-service
        default-for-task-types:
          container: container
          sidecar: sidecar
          container_array: k8s-array
          chatgpt: agent-service
    plugins:
      agent-service:
        supportedTaskTypes:
        - chatgpt
        # Configuring the timeout is optional.
        # Tasks like using ChatGPT with a large model might require a longer time,
        # so we have the option to adjust the timeout setting here.
        defaultAgent:
          timeouts:
            ExecuteTaskSync: 10s

Add the OpenAI API token¶

Install flyteagent pod using helm:

helm repo add flyteorg https://flyteorg.github.io/flyte
helm install flyteagent flyteorg/flyteagent --namespace flyte

Get the base64 value of your OpenAI API token:

echo -n "<OPENAI_API_TOKEN>" | base64

Edit the flyteagent secret:

kubectl edit secret flyteagent -n flyte

apiVersion: v1
data:
  flyte_openai_api_key: <BASE64_ENCODED_OPENAI_API_TOKEN>
kind: Secret
metadata:
  annotations:
    meta.helm.sh/release-name: flyteagent
    meta.helm.sh/release-namespace: flyte
  creationTimestamp: "2023-10-04T04:09:03Z"
  labels:
    app.kubernetes.io/managed-by: Helm
  name: flyteagent
  namespace: flyte
  resourceVersion: "753"
  uid: 5ac1e1b6-2a4c-4e26-9001-d4ba72c39e54
type: Opaque

Upgrade the Flyte Helm release¶

helm upgrade <RELEASE_NAME> flyteorg/flyte-binary -n <YOUR_NAMESPACE> --values <YOUR_YAML_FILE>

Replace <RELEASE_NAME> with the name of your release (e.g., flyte-backend), <YOUR_NAMESPACE> with the name of your namespace (e.g., flyte), and <YOUR_YAML_FILE> with the name of your YAML file.

helm upgrade <RELEASE_NAME> flyte/flyte-core -n <YOUR_NAMESPACE> --values values-override.yaml

Replace <RELEASE_NAME> with the name of your release (e.g., flyte)

and <YOUR_NAMESPACE> with the name of your namespace (e.g., flyte).

For ChatGPT agent on the Flyte cluster, see ChatGPT agent.