# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand. overlay: 1.0.0 info: title: API Evangelist conversational phrasing for Hugging Face Inference Endpoints API version: 1.0.0 extends: openapi/hugging-face-endpoints-api-openapi.yml actions: - target: $.info update: x-apievangelist-phrasing: method: generated generated: '2026-10-01' generator: build-phrasing.py label: Generated by API Evangelist operations: 10 - target: $.paths['/endpoint/{namespace}'].get update: x-apievangelist-phrasing: intent: List Inference Endpoints in a namespace effect: read questions: - Which Inference Endpoints are running under my account or organization? - Can I see every dedicated endpoint my team has deployed? instructions: - text: List all Inference Endpoints in namespace {namespace}. slots: namespace: path.namespace - text: Show me the endpoints my organization {namespace} has deployed. slots: namespace: path.namespace method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}'].post update: x-apievangelist-phrasing: intent: Deploy a new Inference Endpoint effect: write questions: - How do I deploy a model as a dedicated Inference Endpoint on Hugging Face? - Can I pick the cloud provider and hardware when I create an endpoint? - Can a new endpoint be made public, protected or private? instructions: - text: Create an Inference Endpoint named {name} in {namespace} serving {model} on {compute} from {provider} with security type {type}. slots: name: requestBody.name namespace: path.namespace model: requestBody.model compute: requestBody.compute provider: requestBody.provider type: requestBody.type - text: Deploy {model} to a new {type} endpoint called {name} under {namespace} using {provider} hardware {compute}. slots: model: requestBody.model type: requestBody.type name: requestBody.name namespace: path.namespace provider: requestBody.provider compute: requestBody.compute method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}/{endpoint_name}'].get update: x-apievangelist-phrasing: intent: Get an Inference Endpoint's details effect: read questions: - What is the current status and URL of one of my Inference Endpoints? - Which model and hardware is a specific endpoint configured with? instructions: - text: Get the details of endpoint {endpoint_name} in {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name - text: Check the status of Inference Endpoint {endpoint_name} owned by {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}/{endpoint_name}'].put update: x-apievangelist-phrasing: intent: Change an endpoint's model, hardware or scaling effect: write questions: - Can I swap the model on an existing endpoint without recreating it? - Is it possible to move a running endpoint to bigger hardware or change its autoscaling? instructions: - text: Update endpoint {endpoint_name} in {namespace} to serve {model}. slots: namespace: path.namespace endpoint_name: path.endpoint_name model: requestBody.model - text: Change the compute settings of {endpoint_name} under {namespace} to {compute}. slots: namespace: path.namespace endpoint_name: path.endpoint_name compute: requestBody.compute - text: Switch the security type of endpoint {endpoint_name} in {namespace} to {type}. slots: namespace: path.namespace endpoint_name: path.endpoint_name type: requestBody.type method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}/{endpoint_name}'].delete update: x-apievangelist-phrasing: intent: Permanently delete an Inference Endpoint effect: destructive questions: - Can I permanently remove an endpoint and everything attached to it? - What happens to an endpoint's resources when I delete it rather than pause it? instructions: - text: Delete Inference Endpoint {endpoint_name} from {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name - text: Permanently tear down endpoint {endpoint_name} in namespace {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}/{endpoint_name}/pause'].post update: x-apievangelist-phrasing: intent: Pause an endpoint to stop compute costs effect: write questions: - Can I stop paying for an endpoint's compute but keep its configuration for later? - Is there a way to put an endpoint on hold until I resume it manually? instructions: - text: Pause endpoint {endpoint_name} in {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name - text: Stop billing on {endpoint_name} under {namespace} by pausing it. slots: namespace: path.namespace endpoint_name: path.endpoint_name method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}/{endpoint_name}/resume'].post update: x-apievangelist-phrasing: intent: Resume a paused endpoint effect: write questions: - How do I bring a paused Inference Endpoint back online? - Can I restart an endpoint I paused earlier with the same configuration? instructions: - text: Resume paused endpoint {endpoint_name} in {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name - text: Bring {endpoint_name} under {namespace} back online. slots: namespace: path.namespace endpoint_name: path.endpoint_name method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}/{endpoint_name}/scale-to-zero'].post update: x-apievangelist-phrasing: intent: Scale an endpoint down to zero replicas effect: write questions: - Can I scale an endpoint to zero replicas so it wakes up again when traffic arrives? - What is the difference between scaling to zero and pausing when autoscaling is on? instructions: - text: Scale endpoint {endpoint_name} in {namespace} to zero replicas. slots: namespace: path.namespace endpoint_name: path.endpoint_name - text: Drop {endpoint_name} under {namespace} to zero and let traffic scale it back up. slots: namespace: path.namespace endpoint_name: path.endpoint_name method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}/{endpoint_name}/logs'].get update: x-apievangelist-phrasing: intent: Read an endpoint's runtime logs effect: read questions: - Where can I see the runtime logs to debug why my endpoint is failing? - Can I pull the container logs of a specific Inference Endpoint? instructions: - text: Get the runtime logs for endpoint {endpoint_name} in {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name - text: Show me recent log output from {endpoint_name} under {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name method: generated generated: '2026-10-01' - target: $.paths['/endpoint/{namespace}/{endpoint_name}/metrics'].get update: x-apievangelist-phrasing: intent: Get an endpoint's traffic and latency metrics effect: read questions: - How many requests is my endpoint handling, and what are its latencies? - Can I check the error rate of an Inference Endpoint? instructions: - text: Get performance metrics for endpoint {endpoint_name} in {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name - text: Show request counts, latency and errors for {endpoint_name} under {namespace}. slots: namespace: path.namespace endpoint_name: path.endpoint_name method: generated generated: '2026-10-01'