--- title: "Get Eval Run Route" method: GET path: "/api/v1/eval/{run_id}" tags: ["eval"] --- # Get Eval Run Route `GET /api/v1/eval/{run_id}` ## Request - Base URL: `https://api.trajectory.ai` (the document declares no server; this is the origin it was published from) - URL: `https://api.trajectory.ai/api/v1/eval/{run_id}` - Auth: none declared ## Path parameters - `run_id` string, required ## Response `200` Successful Response - EvalRun - `task_selection` EvalTaskSelection - `split` 'test' | 'all', required - `available_task_count` integer, required: Tasks in the selected scope at launch. - `selected_task_count` integer, required: Distinct tasks selected for execution. - `max_samples` integer, nullable: Explicit options.val_num_tasks, or null to run every task in the selected scope. - `agent_id` string, nullable, required: Persisted owner; null for unassigned legacy runs. - `eval_run_id` string, required: Unique evaluation run identifier. - `bench_id` string, required: Benchmark evaluated by this run. - `rollout_xid` string, nullable: XID containing this run's evaluation trajectories. - `display_name` string, nullable, required: User-facing name for the evaluation run; null for legacy rows. - `base_model_slug` string, nullable: Base model evaluated by this run. - `options` RecordedRunOptions, required: Recorded settings retain values outside the public launch limits. - `load_weights_only` boolean, nullable - `density_aware_training` boolean, nullable - `algorithm` 'gspo' | 'grpo' - `train_on_truncation` boolean, nullable - `train_on_error` boolean, nullable - `train_on_timeout` boolean, nullable - `val_enabled` boolean, nullable - `val_interval_steps` integer, nullable - `val_num_tasks` integer, nullable - `val_trajectories_per_task` integer, nullable - `val_max_concurrent_trajectories` integer, nullable - `val_final_wait_timeout_seconds` number, nullable - `learning_rate` number, nullable - `lr_warmup_steps` integer, nullable - `lr_hold_steps` integer, nullable - `lr_min_ratio` number, nullable - `reasoning_effort` 'off' | 'on' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' | 'max', nullable - `temperature` number, nullable - `top_p` number, nullable - `top_k` integer, nullable - `val_reasoning_effort` 'off' | 'on' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' | 'max', nullable - `val_temperature` number, nullable - `val_top_p` number, nullable - `val_top_k` integer, nullable - `val_max_output_tokens_per_turn` integer, nullable - `val_max_output_tokens_per_trajectory` integer, nullable - `val_max_turns_per_trajectory` integer, nullable - `val_trajectory_timeout_seconds` integer, nullable - `num_steps` integer, nullable - `batch_size` integer, nullable - `group_size` integer, nullable - `min_trajectories_per_group` integer, nullable - `max_concurrent_trajectories` integer, nullable - `max_staleness` integer, nullable - `train_on_all_fail` boolean, nullable - `train_on_all_pass` boolean, nullable - `train_on_zero_advantage` boolean, nullable - `trajectory_timeout_seconds` integer, nullable - `batch_timeout_seconds` integer, nullable - `max_output_tokens_per_turn` integer, nullable - `max_output_tokens_per_trajectory` integer, nullable - `max_turns_per_trajectory` integer, nullable - `status` 'pending' | 'registered' | 'running' | 'completed' | 'failed' | 'cancelled', required - `deployment_id` string, nullable: Hosted deployment snapshotted at launch. - `parent_checkpoint_id` string, nullable: Checkpoint evaluated by this run, when applicable. - `training_run_id` string, nullable: Training run that produced the evaluated checkpoint. - `step` integer, nullable: Evaluated checkpoint training step. - `reward_mean` number, nullable: Completed mean of effective evaluation rewards; cut-off trajectories can contribute zero without a logged grader reward. - `total_rollouts` integer, nullable: Rollout total fixed at launch; null for legacy or diagnostic runs. - `created_at` string, date-time, required: Time when the evaluation run was created. - `updated_at` string, date-time, required: Time when the evaluation run was last updated. - `completed_at` string, date-time, nullable: Time when the evaluation reached a terminal status. - `failure` AsyncFailure - `type` string, required: Stable category used for exception mapping. - `code` string, required: Stable programmatic reason for the asynchronous failure. - `message` string, required: Human-readable explanation for this failure occurrence. - `context` object: Machine-readable values specific to this failure code. ## Changes - **2026-11-02** `5b5a8ae4b4a4`: 4 info - added the optional property `options/val_max_output_tokens_per_trajectory` to the response with the `200` status - added the optional property `options/val_max_output_tokens_per_turn` to the response with the `200` status - added the optional property `options/val_max_turns_per_trajectory` to the response with the `200` status - added the optional property `options/val_trajectory_timeout_seconds` to the response with the `200` status - **2026-11-02** `ac63eff4913f`: 21 warning, 24 info - removed the optional property `options/batch_collection_idle_timeout_seconds` from the response with the `200` status - removed the optional property `options/evaluation_collection_timeout_seconds` from the response with the `200` status - removed the optional property `options/evaluation_enabled` from the response with the `200` status - removed the optional property `options/evaluation_frequency` from the response with the `200` status - …41 more - **2026-10-31** `7b218d495d5c`: 1 info - added the optional property `options/load_weights_only` to the response with the `200` status - **2026-10-23** `0a17f8723be0`: 2 info - added the optional property `options/group_size` to the response with the `200` status - added the optional property `options/min_samples_per_group` to the response with the `200` status - …earlier changes not shown [Operation history](https://skmtc.dev/trajectory/apis/trajectory-api/changes/api/v1/eval/:run_id/get.md) --- [API](https://skmtc.dev/trajectory/apis/trajectory-api.md) - [All operations](https://skmtc.dev/trajectory/apis/trajectory-api/llms.txt) - [OpenAPI document](https://skmtc.dev/trajectory/apis/trajectory-api/revisions/5b5a8ae4b4a4?raw)