In Prefect 3.x, the YAML goes under a deployment’s

work_pool.job_variables

, but the exact keys depend on the worker type. For Kubernetes work pools, use request/limit fields:

yaml
deployments:
  - name: my-deployment
    entrypoint: flows/my_flow.py:my_flow
    work_pool:
      name: my-k8s-pool
      job_variables:
        cpu_request: "500m"
        cpu_limit: "1"
        memory_request: "512Mi"
        memory_limit: "1Gi"

For ECS work pools, use

cpu

and `memory`:

yaml
deployments:
  - name: my-deployment
    entrypoint: flows/my_flow.py:my_flow
    work_pool:
      name: my-ecs-pool
      job_variables:
        cpu: 1024
        memory: 2048

A few important caveats: - These keys must exist in the work pool’s base job template or the override won’t do anything. - The valid names/formats are worker-specific: - Kubernetes:

cpu_request

,

cpu_limit

,

memory_request

,

memory_limit

- ECS:

cpu

,

memory

- Kubernetes values use K8s resource syntax like

"500m"

or

"1Gi"

. If you want to verify what variables your pool supports, inspect the default/base template for that worker type. The relevant CLI is:

prefect work-pool get-default-base-job-template --type <worker-type>

Docs: - Work pools and job variables - prefect.yaml reference If you tell me which work pool type you’re using (

kubernetes

,

ecs

,

docker

, etc.), I can give you the exact YAML snippet for that pool.