--- title: "Create a rollout" method: POST path: "/v2/projects/{projectId}/endpoints/{endpointId}/rollouts" tags: ["RolloutService"] --- # Create a rollout `POST /v2/projects/{projectId}/endpoints/{endpointId}/rollouts` ## Request - Base URL: `https://api.together.ai` - URL: `https://api.together.ai/v2/projects/{projectId}/endpoints/{endpointId}/rollouts` - Auth: HTTP bearer Creates a rollout in the pending state without shifting traffic. Start the rollout in a separate request after reviewing its strategy and metric gates. ## Path parameters - `projectId` string, required: Project identifier. - `endpointId` string, required: Endpoint identifier. ## Request body - union: Strategy, metric gates, timing, and cleanup policy for shifting traffic between two deployments under one endpoint. - object - `sourceDeploymentId` string, required: Deployment that traffic shifts away from. - `targetDeploymentId` string, required: Deployment that traffic shifts toward. - `canary` DECanaryConfig, required: Canary strategy configuration for gradual traffic progression. An empty config uses the default 5, 25, 50, 100 percent ladder; over a frozen traffic-split pair left by cancel, the default ladder is derived at start from the pair's current served share so it begins above it. - `steps` DERolloutStep[]: Optional progression steps. Defaults to 5, 25, 50, 100 percent when empty; explicit steps must increase and end at 100 percent. - `traffic` integer, required: Required percentage of traffic on the target deployment for this step. - `replicas` integer: Optional explicit target replica count for this step. - `stepInterval` string: Optional positive soak between steps. Defaults to 3m if omitted, and grows to cover metric rule windows plus ingestion lag. - `blueGreen` DEBlueGreenConfig: Blue-green strategy configuration for a single cutover to the target deployment. - `rolling` DERollingConfig: Rolling strategy configuration for small batches that ramp target replicas up while shrinking source replicas to what their remaining traffic share needs. - `metrics` DEMetricRule[]: Optional metric gates evaluated after each step's soak. Canary only; rejected on rolling and blue-green rollouts. - union: Metric gate evaluated during a rollout. - object - `name` 'inflight_requests' | 'router_error_rate' | 'router_latency', required: Required catalogue key for the metric to gate on. `serving_latency` is retired. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. Optional for router_error_rate and inflight_requests; omitted values default to METRIC_STAT_TYPE_AVG. Required for router_latency, where AVG or PERCENTILE may be used. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `thresholdCheck` DEThresholdCheck, required: Threshold criteria that fail when the target metric violates the configured bound. - `value` number: Finite threshold value. Interpreted in the metric's unit: router_error_rate is a ratio in [0, 1], router_latency is milliseconds, and inflight_requests is in-flight requests per ready replica averaged over the rule window. Thresholds that no achievable value could pass, or that every achievable value passes, are rejected at create. Omitting this value is read as 0. Set 0 explicitly for the strictest threshold: nothing at all is tolerated. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE', required: Required comparison operator applied to the target metric value. - `regressionCheck` DERegressionCheck: Regression criteria that fail when the target regresses against the source beyond a limit. - `maxRegressionPercent` number: Finite maximum allowed regression percentage, greater than or equal to 0. Omitting this value is read as 0. A value of 0 is the strictest budget; any regression fails, and exactly-at-budget passes. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE', required: Required direction that indicates whether higher or lower metric values are worse. - `window` string: Optional query window for the metric. Defaults to the step soak duration. - object - `name` 'inflight_requests' | 'router_error_rate' | 'router_latency', required: Required catalogue key for the metric to gate on. `serving_latency` is retired. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. Optional for router_error_rate and inflight_requests; omitted values default to METRIC_STAT_TYPE_AVG. Required for router_latency, where AVG or PERCENTILE may be used. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `thresholdCheck` DEThresholdCheck: Threshold criteria that fail when the target metric violates the configured bound. - `value` number: Finite threshold value. Interpreted in the metric's unit: router_error_rate is a ratio in [0, 1], router_latency is milliseconds, and inflight_requests is in-flight requests per ready replica averaged over the rule window. Thresholds that no achievable value could pass, or that every achievable value passes, are rejected at create. Omitting this value is read as 0. Set 0 explicitly for the strictest threshold: nothing at all is tolerated. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE', required: Required comparison operator applied to the target metric value. - `regressionCheck` DERegressionCheck, required: Regression criteria that fail when the target regresses against the source beyond a limit. - `maxRegressionPercent` number: Finite maximum allowed regression percentage, greater than or equal to 0. Omitting this value is read as 0. A value of 0 is the strictest budget; any regression fails, and exactly-at-budget passes. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE', required: Required direction that indicates whether higher or lower metric values are worse. - `window` string: Optional query window for the metric. Defaults to the step soak duration. - `finalSourceReplicas` integer: Optional final replica count for the source deployment. Defaults to 0, which drains and stops the source. - `finalTargetReplicas` integer: Optional target replica floor at completion. Must be at least 1 when set; defaults to the source deployment's replica count at create time, or to the source and target deployments' combined replica count when both already stand in the endpoint traffic split after a cancel. The completed target's autoscaling max lands at the landing ceiling, max(this value, the source max, the target's own max); the rollout may lift the target max at first wake, at the first step that needs it, or at completion unless an operator changes max mid-run. The lifted ceiling remains after completion, and PreviewRolloutDefaults reports a coming lift as ROLLOUT_WILL_RAISE_TARGET_MAX. A pre-existing target whose own autoscaling min is higher keeps that floor, reported as FINAL_BELOW_INHERITED_MIN. A target that starts stopped lands exactly at this value; if the source min was higher, PreviewRolloutDefaults reports FINAL_BELOW_SOURCE_MIN. - object - `sourceDeploymentId` string, required: Deployment that traffic shifts away from. - `targetDeploymentId` string, required: Deployment that traffic shifts toward. - `canary` DECanaryConfig: Canary strategy configuration for gradual traffic progression. An empty config uses the default 5, 25, 50, 100 percent ladder; over a frozen traffic-split pair left by cancel, the default ladder is derived at start from the pair's current served share so it begins above it. - `steps` DERolloutStep[]: Optional progression steps. Defaults to 5, 25, 50, 100 percent when empty; explicit steps must increase and end at 100 percent. - `traffic` integer, required: Required percentage of traffic on the target deployment for this step. - `replicas` integer: Optional explicit target replica count for this step. - `stepInterval` string: Optional positive soak between steps. Defaults to 3m if omitted, and grows to cover metric rule windows plus ingestion lag. - `blueGreen` DEBlueGreenConfig, required: Blue-green strategy configuration for a single cutover to the target deployment. - `rolling` DERollingConfig: Rolling strategy configuration for small batches that ramp target replicas up while shrinking source replicas to what their remaining traffic share needs. - `metrics` DEMetricRule[]: Optional metric gates evaluated after each step's soak. Canary only; rejected on rolling and blue-green rollouts. - union: Metric gate evaluated during a rollout. - object - `name` 'inflight_requests' | 'router_error_rate' | 'router_latency', required: Required catalogue key for the metric to gate on. `serving_latency` is retired. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. Optional for router_error_rate and inflight_requests; omitted values default to METRIC_STAT_TYPE_AVG. Required for router_latency, where AVG or PERCENTILE may be used. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `thresholdCheck` DEThresholdCheck, required: Threshold criteria that fail when the target metric violates the configured bound. - `value` number: Finite threshold value. Interpreted in the metric's unit: router_error_rate is a ratio in [0, 1], router_latency is milliseconds, and inflight_requests is in-flight requests per ready replica averaged over the rule window. Thresholds that no achievable value could pass, or that every achievable value passes, are rejected at create. Omitting this value is read as 0. Set 0 explicitly for the strictest threshold: nothing at all is tolerated. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE', required: Required comparison operator applied to the target metric value. - `regressionCheck` DERegressionCheck: Regression criteria that fail when the target regresses against the source beyond a limit. - `maxRegressionPercent` number: Finite maximum allowed regression percentage, greater than or equal to 0. Omitting this value is read as 0. A value of 0 is the strictest budget; any regression fails, and exactly-at-budget passes. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE', required: Required direction that indicates whether higher or lower metric values are worse. - `window` string: Optional query window for the metric. Defaults to the step soak duration. - object - `name` 'inflight_requests' | 'router_error_rate' | 'router_latency', required: Required catalogue key for the metric to gate on. `serving_latency` is retired. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. Optional for router_error_rate and inflight_requests; omitted values default to METRIC_STAT_TYPE_AVG. Required for router_latency, where AVG or PERCENTILE may be used. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `thresholdCheck` DEThresholdCheck: Threshold criteria that fail when the target metric violates the configured bound. - `value` number: Finite threshold value. Interpreted in the metric's unit: router_error_rate is a ratio in [0, 1], router_latency is milliseconds, and inflight_requests is in-flight requests per ready replica averaged over the rule window. Thresholds that no achievable value could pass, or that every achievable value passes, are rejected at create. Omitting this value is read as 0. Set 0 explicitly for the strictest threshold: nothing at all is tolerated. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE', required: Required comparison operator applied to the target metric value. - `regressionCheck` DERegressionCheck, required: Regression criteria that fail when the target regresses against the source beyond a limit. - `maxRegressionPercent` number: Finite maximum allowed regression percentage, greater than or equal to 0. Omitting this value is read as 0. A value of 0 is the strictest budget; any regression fails, and exactly-at-budget passes. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE', required: Required direction that indicates whether higher or lower metric values are worse. - `window` string: Optional query window for the metric. Defaults to the step soak duration. - `finalSourceReplicas` integer: Optional final replica count for the source deployment. Defaults to 0, which drains and stops the source. - `finalTargetReplicas` integer: Optional target replica floor at completion. Must be at least 1 when set; defaults to the source deployment's replica count at create time, or to the source and target deployments' combined replica count when both already stand in the endpoint traffic split after a cancel. The completed target's autoscaling max lands at the landing ceiling, max(this value, the source max, the target's own max); the rollout may lift the target max at first wake, at the first step that needs it, or at completion unless an operator changes max mid-run. The lifted ceiling remains after completion, and PreviewRolloutDefaults reports a coming lift as ROLLOUT_WILL_RAISE_TARGET_MAX. A pre-existing target whose own autoscaling min is higher keeps that floor, reported as FINAL_BELOW_INHERITED_MIN. A target that starts stopped lands exactly at this value; if the source min was higher, PreviewRolloutDefaults reports FINAL_BELOW_SOURCE_MIN. - object - `sourceDeploymentId` string, required: Deployment that traffic shifts away from. - `targetDeploymentId` string, required: Deployment that traffic shifts toward. - `canary` DECanaryConfig: Canary strategy configuration for gradual traffic progression. An empty config uses the default 5, 25, 50, 100 percent ladder; over a frozen traffic-split pair left by cancel, the default ladder is derived at start from the pair's current served share so it begins above it. - `steps` DERolloutStep[]: Optional progression steps. Defaults to 5, 25, 50, 100 percent when empty; explicit steps must increase and end at 100 percent. - `traffic` integer, required: Required percentage of traffic on the target deployment for this step. - `replicas` integer: Optional explicit target replica count for this step. - `stepInterval` string: Optional positive soak between steps. Defaults to 3m if omitted, and grows to cover metric rule windows plus ingestion lag. - `blueGreen` DEBlueGreenConfig: Blue-green strategy configuration for a single cutover to the target deployment. - `rolling` DERollingConfig, required: Rolling strategy configuration for small batches that ramp target replicas up while shrinking source replicas to what their remaining traffic share needs. - `metrics` DEMetricRule[]: Optional metric gates evaluated after each step's soak. Canary only; rejected on rolling and blue-green rollouts. - union: Metric gate evaluated during a rollout. - object - `name` 'inflight_requests' | 'router_error_rate' | 'router_latency', required: Required catalogue key for the metric to gate on. `serving_latency` is retired. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. Optional for router_error_rate and inflight_requests; omitted values default to METRIC_STAT_TYPE_AVG. Required for router_latency, where AVG or PERCENTILE may be used. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `thresholdCheck` DEThresholdCheck, required: Threshold criteria that fail when the target metric violates the configured bound. - `value` number: Finite threshold value. Interpreted in the metric's unit: router_error_rate is a ratio in [0, 1], router_latency is milliseconds, and inflight_requests is in-flight requests per ready replica averaged over the rule window. Thresholds that no achievable value could pass, or that every achievable value passes, are rejected at create. Omitting this value is read as 0. Set 0 explicitly for the strictest threshold: nothing at all is tolerated. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE', required: Required comparison operator applied to the target metric value. - `regressionCheck` DERegressionCheck: Regression criteria that fail when the target regresses against the source beyond a limit. - `maxRegressionPercent` number: Finite maximum allowed regression percentage, greater than or equal to 0. Omitting this value is read as 0. A value of 0 is the strictest budget; any regression fails, and exactly-at-budget passes. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE', required: Required direction that indicates whether higher or lower metric values are worse. - `window` string: Optional query window for the metric. Defaults to the step soak duration. - object - `name` 'inflight_requests' | 'router_error_rate' | 'router_latency', required: Required catalogue key for the metric to gate on. `serving_latency` is retired. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. Optional for router_error_rate and inflight_requests; omitted values default to METRIC_STAT_TYPE_AVG. Required for router_latency, where AVG or PERCENTILE may be used. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `thresholdCheck` DEThresholdCheck: Threshold criteria that fail when the target metric violates the configured bound. - `value` number: Finite threshold value. Interpreted in the metric's unit: router_error_rate is a ratio in [0, 1], router_latency is milliseconds, and inflight_requests is in-flight requests per ready replica averaged over the rule window. Thresholds that no achievable value could pass, or that every achievable value passes, are rejected at create. Omitting this value is read as 0. Set 0 explicitly for the strictest threshold: nothing at all is tolerated. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE', required: Required comparison operator applied to the target metric value. - `regressionCheck` DERegressionCheck, required: Regression criteria that fail when the target regresses against the source beyond a limit. - `maxRegressionPercent` number: Finite maximum allowed regression percentage, greater than or equal to 0. Omitting this value is read as 0. A value of 0 is the strictest budget; any regression fails, and exactly-at-budget passes. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE', required: Required direction that indicates whether higher or lower metric values are worse. - `window` string: Optional query window for the metric. Defaults to the step soak duration. - `finalSourceReplicas` integer: Optional final replica count for the source deployment. Defaults to 0, which drains and stops the source. - `finalTargetReplicas` integer: Optional target replica floor at completion. Must be at least 1 when set; defaults to the source deployment's replica count at create time, or to the source and target deployments' combined replica count when both already stand in the endpoint traffic split after a cancel. The completed target's autoscaling max lands at the landing ceiling, max(this value, the source max, the target's own max); the rollout may lift the target max at first wake, at the first step that needs it, or at completion unless an operator changes max mid-run. The lifted ceiling remains after completion, and PreviewRolloutDefaults reports a coming lift as ROLLOUT_WILL_RAISE_TARGET_MAX. A pre-existing target whose own autoscaling min is higher keeps that floor, reported as FINAL_BELOW_INHERITED_MIN. A target that starts stopped lands exactly at this value; if the source min was higher, PreviewRolloutDefaults reports FINAL_BELOW_SOURCE_MIN. ## Response `200` OK - DERollout: Public view of a rollout resource, including runtime progress and any pause or abort reason. - `id` string, required: Output only. Unique rollout identifier. - `endpointId` string, required: Output only. Endpoint this rollout belongs to. - `sourceDeploymentId` string, required: Output only. Deployment that traffic is shifting away from. - `targetDeploymentId` string, required: Output only. Deployment that traffic is shifting toward. - `strategy` 'ROLLOUT_STRATEGY_TYPE_ROLLING' | 'ROLLOUT_STRATEGY_TYPE_CANARY' | 'ROLLOUT_STRATEGY_TYPE_BLUE_GREEN', required: Output only. Rollout strategy selected at creation. - `state` 'ROLLOUT_STATE_RUNNING' | 'ROLLOUT_STATE_PAUSED' | 'ROLLOUT_STATE_STABILIZING' | 'ROLLOUT_STATE_COMPLETED' | 'ROLLOUT_STATE_PENDING' | 'ROLLOUT_STATE_SYSTEM_PAUSED' | 'ROLLOUT_STATE_CANCELLING' | 'ROLLOUT_STATE_CANCELED' | 'ROLLOUT_STATE_PAUSING', required: Output only. High-level rollout lifecycle state. - `currentStep` integer: Output only. Zero-based index of the current step. Unset while PENDING; step 0 is reported explicitly after start. - `currentTrafficPercent` integer: Output only. Applied percentage of traffic on the target deployment. - `createdAt` string, date-time, required: Output only. Timestamp when the rollout was created. - `startedAt` string, date-time: Output only. Timestamp when the rollout started running. - `completedAt` string, date-time: Output only. Timestamp when the rollout reached a terminal state. - `etag` string: Output only. Opaque version tag for optimistic concurrency control. - `status` DERolloutStatus, required: Derived runtime progress for a rollout. - `totalSteps` integer, required: Total number of steps in the rollout progression. Always serializes when status is present. - `condition` DERolloutCondition: Structured reason a rollout stopped progressing. - `category` 'ROLLOUT_FAILURE_CATEGORY_METRIC_REGRESSION' | 'ROLLOUT_FAILURE_CATEGORY_METRICS_UNAVAILABLE' | 'ROLLOUT_FAILURE_CATEGORY_TARGET_NOT_READY' | 'ROLLOUT_FAILURE_CATEGORY_SOURCE_NOT_DRAINED' | 'ROLLOUT_FAILURE_CATEGORY_HEALTH_REGRESSION' | 'ROLLOUT_FAILURE_CATEGORY_CAPACITY_EXHAUSTED' | 'ROLLOUT_FAILURE_CATEGORY_ROUTING_ERROR' | 'ROLLOUT_FAILURE_CATEGORY_DEPENDENCY_OUTAGE' | 'ROLLOUT_FAILURE_CATEGORY_ABORTED_BY_OPERATOR' | 'ROLLOUT_FAILURE_CATEGORY_INTERNAL' | 'ROLLOUT_FAILURE_CATEGORY_POLICY_INFEASIBLE' | 'ROLLOUT_FAILURE_CATEGORY_UNDER_SERVED' | 'ROLLOUT_FAILURE_CATEGORY_ENTITLEMENT_LAPSED': Category that classifies why the rollout stopped. - `message` string: Human-readable explanation for the condition. - `atStep` integer: Step index where the condition arose. Step 0 serializes explicitly. - `observedAt` string, date-time: Timestamp when the condition was observed. - `metrics` DEMetricResult[]: Metrics observed at the failing gate, enriched with their criteria. Unmeasured rules appear as synthesized rows with verdict METRIC_VERDICT_UNAVAILABLE and no measured values. - `name` string: Metric name as exported to the observability backend. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `check` 'METRIC_CHECK_TYPE_THRESHOLD' | 'METRIC_CHECK_TYPE_REGRESSION': Evaluation form used by the metric rule. - `sourceValue` number: Observed source baseline. Set only for regression checks with a recorded observation; a 0 reading serializes explicitly. - `targetValue` number: Observed target value. Set when the gate recorded an observation; absent on synthesized unavailable results. A 0 reading serializes explicitly. - `threshold` number: Threshold criteria used when check is METRIC_CHECK_TYPE_THRESHOLD. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE': Threshold comparison operator. - `maxRegressionPercent` number: Regression percentage limit used when check is METRIC_CHECK_TYPE_REGRESSION. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE': Direction that indicates whether higher or lower values are worse. - `verdict` 'METRIC_VERDICT_PASS' | 'METRIC_VERDICT_BREACHED' | 'METRIC_VERDICT_UNAVAILABLE': Rule decision recorded by the metric gate. Absent when no decision was recorded. - `failureReason` string: Rule-specific failure text. Set only when verdict is METRIC_VERDICT_BREACHED and the gate recorded one. - `type` 'CapacityLimited': Informational condition type. `CapacityLimited` means the current step advanced partially because full capacity was not placeable. - `steps` DERolloutStepStatus[], required: Per-step rollout execution summaries. - `stepIndex` integer: Index of this step in the rollout progression. Step 0 serializes explicitly. - `targetTrafficPercent` integer: Target traffic percentage configured for this step. Always serializes for recorded steps. - `state` 'ROLLOUT_STEP_STATE_PENDING' | 'ROLLOUT_STEP_STATE_RUNNING' | 'ROLLOUT_STEP_STATE_PASSED' | 'ROLLOUT_STEP_STATE_FAILED' | 'ROLLOUT_STEP_STATE_PAUSED' | 'ROLLOUT_STEP_STATE_CANCELED' | 'ROLLOUT_STEP_STATE_SKIPPED': Outcome of this step. Finished steps are PASSED, the live step mirrors the rollout state, skipped-over steps are SKIPPED, and unreached steps are PENDING. - `metrics` DEMetricResult[]: Metric gate results for this step, enriched with criteria and verdict. Unmeasured rules appear as synthesized rows with verdict METRIC_VERDICT_UNAVAILABLE and no measured values. - `name` string: Metric name as exported to the observability backend. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `check` 'METRIC_CHECK_TYPE_THRESHOLD' | 'METRIC_CHECK_TYPE_REGRESSION': Evaluation form used by the metric rule. - `sourceValue` number: Observed source baseline. Set only for regression checks with a recorded observation; a 0 reading serializes explicitly. - `targetValue` number: Observed target value. Set when the gate recorded an observation; absent on synthesized unavailable results. A 0 reading serializes explicitly. - `threshold` number: Threshold criteria used when check is METRIC_CHECK_TYPE_THRESHOLD. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE': Threshold comparison operator. - `maxRegressionPercent` number: Regression percentage limit used when check is METRIC_CHECK_TYPE_REGRESSION. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE': Direction that indicates whether higher or lower values are worse. - `verdict` 'METRIC_VERDICT_PASS' | 'METRIC_VERDICT_BREACHED' | 'METRIC_VERDICT_UNAVAILABLE': Rule decision recorded by the metric gate. Absent when no decision was recorded. - `failureReason` string: Rule-specific failure text. Set only when verdict is METRIC_VERDICT_BREACHED and the gate recorded one. - `startedAt` string, date-time: Timestamp when this step's first sub-step ran. Unset for steps no sub-step reached. - `completedAt` string, date-time: Timestamp when this step finished, was skipped over, or the rollout ended on it. Unset while in progress. - `failureReason` string: Failure reason set only when this step failed. - `updatedAt` string, date-time: Timestamp of the most recent progress update. - `conditions` DERolloutCondition[]: Informational conditions that describe the rollout's current state. Omitted when empty; clients should treat an absent key as an empty list. - `category` 'ROLLOUT_FAILURE_CATEGORY_METRIC_REGRESSION' | 'ROLLOUT_FAILURE_CATEGORY_METRICS_UNAVAILABLE' | 'ROLLOUT_FAILURE_CATEGORY_TARGET_NOT_READY' | 'ROLLOUT_FAILURE_CATEGORY_SOURCE_NOT_DRAINED' | 'ROLLOUT_FAILURE_CATEGORY_HEALTH_REGRESSION' | 'ROLLOUT_FAILURE_CATEGORY_CAPACITY_EXHAUSTED' | 'ROLLOUT_FAILURE_CATEGORY_ROUTING_ERROR' | 'ROLLOUT_FAILURE_CATEGORY_DEPENDENCY_OUTAGE' | 'ROLLOUT_FAILURE_CATEGORY_ABORTED_BY_OPERATOR' | 'ROLLOUT_FAILURE_CATEGORY_INTERNAL' | 'ROLLOUT_FAILURE_CATEGORY_POLICY_INFEASIBLE' | 'ROLLOUT_FAILURE_CATEGORY_UNDER_SERVED' | 'ROLLOUT_FAILURE_CATEGORY_ENTITLEMENT_LAPSED': Category that classifies why the rollout stopped. - `message` string: Human-readable explanation for the condition. - `atStep` integer: Step index where the condition arose. Step 0 serializes explicitly. - `observedAt` string, date-time: Timestamp when the condition was observed. - `metrics` DEMetricResult[]: Metrics observed at the failing gate, enriched with their criteria. Unmeasured rules appear as synthesized rows with verdict METRIC_VERDICT_UNAVAILABLE and no measured values. - `name` string: Metric name as exported to the observability backend. - `stat` 'METRIC_STAT_TYPE_AVG' | 'METRIC_STAT_TYPE_PERCENTILE': Aggregation used for the metric. - `percentile` integer: Percentile value, such as 99. Set only when stat is METRIC_STAT_TYPE_PERCENTILE. - `check` 'METRIC_CHECK_TYPE_THRESHOLD' | 'METRIC_CHECK_TYPE_REGRESSION': Evaluation form used by the metric rule. - `sourceValue` number: Observed source baseline. Set only for regression checks with a recorded observation; a 0 reading serializes explicitly. - `targetValue` number: Observed target value. Set when the gate recorded an observation; absent on synthesized unavailable results. A 0 reading serializes explicitly. - `threshold` number: Threshold criteria used when check is METRIC_CHECK_TYPE_THRESHOLD. - `operator` 'THRESHOLD_OPERATOR_GT' | 'THRESHOLD_OPERATOR_GTE' | 'THRESHOLD_OPERATOR_LT' | 'THRESHOLD_OPERATOR_LTE': Threshold comparison operator. - `maxRegressionPercent` number: Regression percentage limit used when check is METRIC_CHECK_TYPE_REGRESSION. - `direction` 'REGRESSION_DIRECTION_HIGHER_IS_WORSE' | 'REGRESSION_DIRECTION_LOWER_IS_WORSE': Direction that indicates whether higher or lower values are worse. - `verdict` 'METRIC_VERDICT_PASS' | 'METRIC_VERDICT_BREACHED' | 'METRIC_VERDICT_UNAVAILABLE': Rule decision recorded by the metric gate. Absent when no decision was recorded. - `failureReason` string: Rule-specific failure text. Set only when verdict is METRIC_VERDICT_BREACHED and the gate recorded one. - `type` 'CapacityLimited': Informational condition type. `CapacityLimited` means the current step advanced partially because full capacity was not placeable. - `pauseInfo` DEPauseInfo: Pause metadata returned while a rollout is paused. - `reason` string: Human-readable reason recorded when the rollout was paused. - `pausedAt` string, date-time, required: Timestamp when the rollout was paused. ## Other responses - `default`: Default error response ## Changes - **2026-10-22** `bb5257b3173a`: 1 info - endpoint added [Operation history](https://skmtc.dev/together/apis/together-apis/changes/v2/projects/:projectId/endpoints/:endpointId/rollouts/post.md) --- [API](https://skmtc.dev/together/apis/together-apis.md) - [All operations](https://skmtc.dev/together/apis/together-apis/llms.txt) - [OpenAPI document](https://skmtc.dev/together/apis/together-apis/revisions/93438ddac711?raw)