> ## Documentation Index > Fetch the complete documentation index at: https://docs.together.ai/llms.txt > Use this file to discover all available pages before exploring further. # Get a deployment > Retrieves a deployment's desired configuration, placement, runtime information, and current provisioning status. ## OpenAPI ````yaml openapi.yaml GET /projects/{projectId}/endpoints/{endpointId}/deployments/{id} openapi: 3.1.0 info: title: Together APIs description: The Together REST API. See https://docs.together.ai for more details. version: 2.0.0 termsOfService: https://www.together.ai/terms-of-service contact: name: Together Support url: https://www.together.ai/contact license: name: MIT url: https://github.com/togethercomputer/openapi/blob/main/LICENSE servers: - url: https://api.together.ai/v1 description: Default environment for APIs - url: https://api-inference.together.ai/v2 description: Optimized environment for inference security: - bearerAuth: [] paths: /projects/{projectId}/endpoints/{endpointId}/deployments/{id}: get: tags: - DeploymentService summary: Get a deployment description: >- Retrieves a deployment's desired configuration, placement, runtime information, and current provisioning status. operationId: DeploymentService_GetDeployment parameters: - name: projectId in: path required: true schema: description: Project identifier. type: string - name: endpointId in: path required: true schema: description: Endpoint identifier. type: string - name: id in: path required: true schema: description: Deployment identifier. type: string responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/DE.Deployment' default: description: Default error response content: application/json: schema: $ref: '#/components/schemas/ErrorData' servers: - url: https://api.together.ai/v2 components: schemas: DE.Deployment: description: >- Serving workload that binds a model and immutable config to an endpoint and manages its replicas. type: object required: - id - projectId - endpointId - name - createdAt - updatedAt - modelId - modelRevisionId - model - autoscaling - configId - config - etag - hardware - trafficMode - status properties: id: readOnly: true type: string description: Unique deployment identifier. projectId: readOnly: true type: string description: ID of the project that owns the deployment. endpointId: type: string description: ID of the endpoint that contains the deployment. name: type: string description: >- Project- and endpoint-qualified deployment name in the form `//`. Pass it as `model` in an inference request to target this deployment directly instead of using the endpoint's traffic split. createdAt: readOnly: true type: string description: Timestamp when the deployment was created. format: date-time updatedAt: readOnly: true type: string description: Timestamp when the deployment was last updated. format: date-time modelId: type: string description: >- Deprecated. Use `model`. Model identifier being served, populated during migration. modelRevisionId: type: string description: >- Deprecated. Use `model` with a /revisions/{revisionId} segment. Pin to a specific model revision. model: type: string description: >- Pinned model resource in the form `projects/{projectId}/models/{modelId}/revisions/{revisionId}`. autoscaling: allOf: - $ref: '#/components/schemas/DE.AutoscalingResponse' description: >- Replica bounds, timing windows, and metrics that control horizontal scaling. configId: type: string description: >- Deprecated. Use `config`. Config revision identifier, populated during migration. config: type: string description: >- Immutable config revision in the form `projects/{projectId}/configs/{configRevisionId}`. speculatorId: readOnly: true type: string description: >- Deprecated. Use `speculator`. Speculative decoding model identifier derived from the deployment config. speculatorRevisionId: readOnly: true type: string description: >- Deprecated. Use `speculator`. ID of the speculative decoding draft-model revision pinned at creation time. speculator: readOnly: true type: string description: >- Pinned draft-model resource used for speculative decoding, in the same form as `model`. Omitted when speculative decoding is disabled. estimatedEffectiveTrafficShare: readOnly: true type: number format: double description: >- Estimated fraction in [0, 1] of endpoint traffic that reaches this deployment under the current routing configuration. Absent or unrouted deployments are 0. maxConcurrentRequestsPerReplica: type: string description: >- Maximum number of inference requests that may be in flight to a single replica. If omitted, the platform uses one less than the config's per-replica concurrency limit to reserve a health-check slot. Values above that maximum are reduced on create and update; 0 means unlimited when the config limit is 1 or less. Changes take effect without restarting replicas. inactiveTimeout: type: integer description: >- Minutes without an inference request before the deployment stops automatically. Omitted or 0 means automatic stopping is disabled. etag: type: string description: >- Opaque version tag for optimistic concurrency control. Supply on update/delete to ensure consistent read-modify-write. If not set, the write overwrites based on current state. hardware: readOnly: true type: string description: >- Hardware selected by the deployment config, including GPU type and count. trafficMode: readOnly: true enum: - TRAFFIC_MODE_LIVE - TRAFFIC_MODE_SHADOW type: string description: >- Whether the deployment serves client-visible responses or only mirrored shadow traffic. runtimeInfo: readOnly: true allOf: - $ref: '#/components/schemas/DE.RuntimeInfo' description: >- Serving engine and feature support derived from the immutable config. desiredReplicas: readOnly: true type: integer description: >- Number of replicas the autoscaler currently wants across all regions. Not settable on any request; steer it through `autoscaling.minReplicas` and `autoscaling.maxReplicas`. status: readOnly: true allOf: - $ref: '#/components/schemas/DE.DeploymentStatus' description: Current lifecycle state and observed replica counts. placement: allOf: - $ref: '#/components/schemas/DE.Placement' description: Region constraints used to schedule the deployment's replicas. ErrorData: type: object required: - error properties: error: type: object properties: message: type: string nullable: false type: type: string nullable: false param: type: string nullable: true default: null code: type: string nullable: true default: null required: - type - message DE.AutoscalingResponse: description: Resolved autoscaling configuration returned for a deployment. allOf: - $ref: '#/components/schemas/DE.Autoscaling' - type: object required: - minReplicas - maxReplicas DE.RuntimeInfo: type: object properties: engineType: type: string description: Serving engine, such as `vllm`, `trtllm`, or `sglang`. engineVersion: type: string description: Version of the serving engine. functionCallingSupported: type: boolean description: Whether the runtime accepts tool and function-calling requests. structuredOutputSupported: type: boolean description: >- Whether the runtime can constrain generation to a structured output schema. description: Runtime information derived from the deployment's configuration. DE.DeploymentStatus: type: object required: - state - message properties: state: enum: - DEPLOYMENT_STATE_PROVISIONING - DEPLOYMENT_STATE_READY - DEPLOYMENT_STATE_SCALING - DEPLOYMENT_STATE_DEGRADED - DEPLOYMENT_STATE_FAILED - DEPLOYMENT_STATE_STOPPED - DEPLOYMENT_STATE_STOPPING type: string description: High-level lifecycle state. readyReplicas: type: integer description: Total replicas actively serving traffic across all clusters. message: type: string description: Human-readable explanation of the current state. scheduledReplicas: type: integer description: Replicas the scheduler has placed on clusters. details: allOf: - $ref: '#/components/schemas/DE.StatusDetails' description: >- Status totals broken down by dimension. Omitted when the breakdown cannot be determined. description: >- Current status of a deployment, derived at read time from internal state. DE.Placement: description: Placement controls where a deployment is scheduled. oneOf: - type: object x-stainless-variantName: Inline required: - inline properties: inline: description: Inline placement parameters evaluated at deploy time. allOf: - $ref: '#/components/schemas/DE.InlinePlacement' - type: object x-stainless-variantName: Profile required: - profile properties: profile: type: string description: UID of a saved placement profile. DE.Autoscaling: type: object properties: minReplicas: type: integer description: >- Minimum number of replicas. Omit on update to preserve the current value. Set both `minReplicas` and `maxReplicas` to `0` to stop the deployment. maxReplicas: type: integer description: >- Maximum number of replicas. Defaults to `minReplicas`; omitting it on update preserves the current value. scaleDownWindow: pattern: ^-?(?:0|[1-9][0-9]{0,11})(?:\.[0-9]{1,9})?s$ type: string description: >- Time a lower replica recommendation must remain stable before scaling down. Defaults to `5m`. scaleUpWindow: pattern: ^-?(?:0|[1-9][0-9]{0,11})(?:\.[0-9]{1,9})?s$ type: string description: Stabilization window before scaling up. scaleToZeroWindow: pattern: ^-?(?:0|[1-9][0-9]{0,11})(?:\.[0-9]{1,9})?s$ type: string description: >- Idle period after which the deployment automatically stops and releases its replicas. scalingMetrics: type: array items: $ref: '#/components/schemas/DE.ScalingMetric' description: >- Metrics and targets that drive replica recommendations. When omitted, the platform uses concurrent in-flight requests per replica. scaleUp: allOf: - $ref: '#/components/schemas/DE.ScalingRules' description: >- Rate limits applied when scaling up. Stabilization remains controlled by `scaleUpWindow`. Omitted fields are preserved on update; a non-empty policy list replaces the previous list. To clear policies or reset the selector, explicitly mask that leaf field. Empty lists in parent-only updates are treated as omitted. scaleDown: allOf: - $ref: '#/components/schemas/DE.ScalingRules' description: >- Rate limits applied when scaling down. Stabilization remains controlled by `scaleDownWindow`. Omitted fields are preserved on update; a non-empty policy list replaces the previous list. To clear policies or reset the selector, explicitly mask that leaf field. Empty lists in parent-only updates are treated as omitted. description: Autoscaling configuration for a deployment. DE.StatusDetails: type: object required: - region properties: region: type: array items: $ref: '#/components/schemas/DE.RegionStatus' description: >- Regions where the deployment is actually scheduled or serving replicas, sorted by region. description: Deployment status broken down by each supported dimension. DE.InlinePlacement: type: object properties: regions: type: array items: type: string description: >- Regions where the deployment is allowed to run. Multiple regions allow best-effort replica spreading. constraint: enum: - ENFORCEMENT_REQUIRED - ENFORCEMENT_PREFERRED type: string description: How strictly the regions list is enforced. compliancePolicy: description: Compliance regimes required for clusters that run the deployment. allOf: - $ref: '#/components/schemas/DE.CompliancePolicy' description: >- Inline placement parameters expanded into scheduling rules by the server. DE.ScalingMetric: type: object properties: name: enum: - active_sessions - cache_hit_rate - decoding_speed - e2e_latency - gpu_utilization - inflight_requests - throughput_per_replica - token_utilization - ttft type: string description: Autoscaling metric name from the server allowlist. type: enum: - METRIC_TARGET_TYPE_VALUE - METRIC_TARGET_TYPE_UTILIZATION - METRIC_TARGET_TYPE_AVERAGE_VALUE type: string description: >- Whether `target` is an absolute value, a utilization percentage, or a per-replica average. target: type: number description: >- Target interpreted according to `type`. Utilization uses a percentage from 0 to 100, value uses an absolute measurement, and average value uses a per-replica measurement. percentile: type: string description: >- Percentile to evaluate for latency-based metrics: `p50`, `p90`, `p95`, or `p99`. description: Metric and target used by the autoscaler to recommend a replica count. required: - name - type - target DE.ScalingRules: type: object properties: policies: type: array items: $ref: '#/components/schemas/DE.ScalingPolicy' description: >- Non-empty lists replace the existing policies. To clear policies, include `autoscaling.scaleDown.policies` or `autoscaling.scaleUp.policies` in the update mask and supply an empty scaling rules object or `policies: []`. selectPolicy: enum: - SCALING_POLICY_SELECT_MAX - SCALING_POLICY_SELECT_MIN - SCALING_POLICY_SELECT_DISABLED type: string description: >- `SCALING_POLICY_SELECT_MIN` chooses the policy allowing the smallest replica change; `SCALING_POLICY_SELECT_MAX` chooses the largest. These are caps, not guaranteed changes. `SCALING_POLICY_SELECT_DISABLED` holds this direction steady while replica bounds still apply. Omitted preserves the existing selector on update. When no selector is configured, authored policies use MAX; with no policies configured, the platform defaults apply. To reset the selector, include `autoscaling.scaleDown.selectPolicy` or `autoscaling.scaleUp.selectPolicy` in the update mask and omit `selectPolicy`. Clear both policies and `selectPolicy` to restore inherited defaults. description: Rate limits applied after stabilization and before replica bounds. example: policies: - type: SCALING_POLICY_TYPE_PERCENT value: 25 periodSeconds: 60 - type: SCALING_POLICY_TYPE_PODS value: 10 periodSeconds: 60 selectPolicy: SCALING_POLICY_SELECT_MIN DE.RegionStatus: type: object required: - region - scheduledReplicas - readyReplicas properties: region: type: string description: >- Region name using the same vocabulary accepted by inline placement regions. scheduledReplicas: type: integer description: Replicas the scheduler has placed in this region. readyReplicas: type: integer description: Replicas serving traffic in this region. description: Realized scheduled and ready replica counts for one deployment region. DE.CompliancePolicy: type: object properties: hipaa: type: boolean description: Restrict placement to HIPAA-attested clusters. description: Compliance regimes required by a deployment placement policy. DE.ScalingPolicy: type: object description: Replica rate-limit policy applied over a trailing window. required: - type - value - periodSeconds properties: type: enum: - SCALING_POLICY_TYPE_PODS - SCALING_POLICY_TYPE_PERCENT type: string description: >- Whether `value` is a replica count or a percentage of the replica count at the start of the trailing period. Scaling events within that period count against the allowance; percentages are rounded to whole replicas. value: type: integer minimum: 1 maximum: 2147483647 description: Positive replica count or percentage used as the rate-limit amount. periodSeconds: type: integer minimum: 1 maximum: 1800 description: Trailing rate-limit window in seconds, from 1 to 1800. securitySchemes: bearerAuth: type: http scheme: bearer x-bearer-format: bearer x-default: default ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.