> ## Documentation Index > Fetch the complete documentation index at: https://docs.trajectory.ai/llms.txt > Use this file to discover all available pages before exploring further. # List training step trajectories > Retrieve a page of one training step’s trajectories, including rewards, termination reasons, tool failures, and group outcomes. The sort presets are errors, mixed, success, and failed. The search parameter is currently ignored. ## OpenAPI ````yaml /openapi.json get /api/v1/train/{run_id}/steps/{step}/trajectories openapi: 3.1.0 info: title: Trajectory API version: 0.1.0 x-trajectory-api-version: '2026-10-14' description: >- This contract is API version 2026-10-14. The optional Trajectory-API-Version header pins it, and tmp-contract is accepted as an alias. Requests without the header get this version. Version 2026-10-07 and the pre-launch legacy response formats are retired. servers: - url: https://api.trajectory.ai security: [] tags: - name: agents x-group: Agents - name: benchmarks x-group: Specs (benchmarks) - name: diagnostics x-group: Diagnostics - name: benchmark-ingestion x-group: Spec (benchmark) submissions - name: eval x-group: Evaluations - name: training x-group: Training - name: deployments x-group: Deployments - name: model-endpoints x-group: Model endpoints - name: inference x-group: Inference - name: trajectories x-group: Trajectories - name: artifacts x-group: Artifacts - name: telemetry x-group: Telemetry - name: supported-models x-group: Supported models - name: organizations x-group: Organizations - name: secrets x-group: Secrets paths: /api/v1/train/{run_id}/steps/{step}/trajectories: get: tags: - training summary: List training step trajectories description: >- Retrieve a page of one training step’s trajectories, including rewards, termination reasons, tool failures, and group outcomes. The sort presets are errors, mixed, success, and failed. The search parameter is currently ignored. operationId: listTrainingStepTrajectories parameters: - name: run_id in: path required: true schema: type: string title: Run Id - name: step in: path required: true schema: type: integer title: Step - name: mode in: query required: false schema: enum: - training - eval type: string default: training title: Mode - name: selection in: query required: false schema: enum: - sampled - trained type: string default: sampled title: Selection - name: cursor in: query required: false schema: anyOf: - type: string - type: 'null' title: Cursor - name: limit in: query required: false schema: type: integer maximum: 200 minimum: 1 default: 25 title: Limit - name: sort in: query required: false schema: anyOf: - type: string - type: 'null' title: Sort - name: order in: query required: false schema: enum: - asc - desc type: string default: desc title: Order - name: search in: query required: false schema: anyOf: - type: string - type: 'null' title: Search responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/TrainingStepTrajectoriesResponse' default: $ref: '#/components/responses/ApiErrorResponse' security: - ApiKeyAuth: [] components: schemas: TrainingStepTrajectoriesResponse: properties: training_run_id: type: string title: Training Run Id description: Training run whose step rollouts are returned. step: type: integer title: Step description: Training step whose rollouts are returned. items: items: $ref: '#/components/schemas/TrainingStepTrajectory' type: array title: Items description: >- Page of the training step's rollouts. Without a sort they are ordered by uid then trajectory_idx; sort=errors puts ERROR terminations then most tool failures first, and sort=mixed|success|failed puts groups with that outcome first. next_cursor: anyOf: - type: string - type: 'null' title: Next Cursor description: Cursor for the next page, or null if none. has_more: type: boolean title: Has More description: True if more rollouts remain beyond this page. type: object required: - training_run_id - step - items - has_more title: TrainingStepTrajectoriesResponse TrainingStepTrajectory: properties: task_preview: anyOf: - type: string - type: 'null' title: Task Preview description: Preview of the task's first user prompt, when recorded. trajectory_id: type: string title: Trajectory Id description: Trajectory identifier. uid: anyOf: - type: string - type: 'null' title: Uid description: 'Rollout group key: rollouts of one task share a uid.' trajectory_idx: anyOf: - type: integer - type: 'null' title: Trajectory Idx description: Position of this rollout within its group. status: anyOf: - $ref: '#/components/schemas/TrajectoryStatus' - type: 'null' description: Recorded trajectory lifecycle status, or null when unavailable. reward: anyOf: - type: number - type: 'null' title: Reward description: Reward assigned to the trajectory, or null until finished. turns: anyOf: - type: integer - type: 'null' title: Turns description: >- Agent turns (num_steps); a turn is an agent step inside a trajectory, distinct from a training step. tokens_generated: anyOf: - type: integer - type: 'null' title: Tokens Generated description: Generated tokens (total_token_count). termination_reason: type: string title: Termination Reason description: Upper-case TerminationReason name; NONE when unset. tool_failure_count: type: integer title: Tool Failure Count description: Tool failures recorded on the trajectory; 0 when null. tool_failures_exec: type: integer title: Tool Failures Exec description: >- Tool calls that ran and failed (attributes.metrics.num_tool_failures_exec). tool_failures_parse: type: integer title: Tool Failures Parse description: Malformed tool calls (attributes.metrics.num_tool_failures_parse). tool_failures_unk: type: integer title: Tool Failures Unk description: >- Tool calls with no response and no error (attributes.metrics.num_tool_failures_unk). training_step: anyOf: - type: integer - type: 'null' title: Training Step description: >- Optimizer step that consumed this rollout; null when not yet trained on. task_id: anyOf: - type: string - type: 'null' title: Task Id description: Benchmark task identifier, when one was recorded. rejection_reason: anyOf: - type: string - type: 'null' title: Rejection Reason description: >- Why training skipped this rollout's group without learning from it: all_pass, all_fail, or no_variance. Reported only for selection=trained; null when trained or still batched. group_outcome: type: string enum: - incomplete - mixed - all_success - all_failed title: Group Outcome description: >- Reward pattern of this rollout's group: incomplete until every rollout has finished with a reward; then all_success (every reward >= 1), all_failed (every reward <= 0), or mixed. Skipped training batches use the trainer's group decision and ignore unscored rollouts. group_no_signal: type: boolean title: Group No Signal description: >- True when every finished reward in the group is identical, so GRPO advantages are zero and the group contributes no gradient whether or not the trainer consumed it. Skipped training batches ignore unscored rollouts; no_variance is authoritative. type: object required: - trajectory_id - uid - trajectory_idx - status - reward - turns - tokens_generated - termination_reason - tool_failure_count - tool_failures_exec - tool_failures_parse - tool_failures_unk - training_step - task_id - rejection_reason - group_outcome - group_no_signal title: TrainingStepTrajectory ApiErrorEnvelope: properties: error: $ref: '#/components/schemas/ApiErrorBody' required: - error title: ApiErrorEnvelope type: object TrajectoryStatus: type: string enum: - in_progress - cancelling - finalizing - completed - aborted title: TrajectoryStatus description: The trajectory’s recorded lifecycle status. ApiErrorBody: properties: type: $ref: '#/components/schemas/ErrorType' code: type: string title: Code message: type: string title: Message param: anyOf: - type: string - type: 'null' default: null title: Param doc_url: type: string title: Doc Url context: additionalProperties: true title: Context type: object required: - type - code - message - doc_url - context title: ApiErrorBody type: object ErrorType: enum: - invalid_request_error - authentication_error - permission_error - conflict_error - billing_error - rate_limit_error - dependency_error - api_error title: ErrorType type: string responses: ApiErrorResponse: description: >- Common API error. Possible codes: `authentication_required`, `permission_denied`, `validation_error`, `rate_limited`, `internal_error`, `service_unavailable`. headers: X-Request-ID: $ref: '#/components/headers/RequestId' Trajectory-Should-Retry: $ref: '#/components/headers/TrajectoryShouldRetry' Retry-After: $ref: '#/components/headers/RetryAfter' content: application/json: schema: $ref: '#/components/schemas/ApiErrorEnvelope' headers: RequestId: schema: type: string TrajectoryShouldRetry: schema: type: boolean RetryAfter: schema: type: string securitySchemes: ApiKeyAuth: type: apiKey in: header name: X-API-Key description: Organization API key ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.