Experiments
List Experiments
Lists experiments in the workspace, newest first.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.list()
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.list();
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"limit": Optional[int],
"starting_after": Optional[str],
"ending_before": Optional[str],
"project_id": Optional[str],
}
{
limit?: number;
startingAfter?: string;
endingBefore?: string;
projectId?: string;
}
Show Response
Show Response
{
"data": [{ # optional
"experiment_id": Optional[str],
"display_name": Optional[str],
"project_id": Optional[str],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"selected_run_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
"source": Optional[str],
"row_height_level": Optional[int],
"unique_evaluators": [{ # optional
"evaluator_id": Optional[str],
"evaluator_type": Optional[str],
"evaluator_name": Optional[str],
"evaluator_key": Optional[str],
}],
"created_by_id": Optional[str],
}],
"has_more": Optional[bool],
}
{
data?: {
experimentId?: string;
displayName?: string;
projectId?: string;
datasetId?: string;
datasetVersionId?: string;
selectedRunId?: string;
createdAt?: Date;
updatedAt?: Date;
source?: string;
rowHeightLevel?: number;
uniqueEvaluators?: {
evaluatorId?: string;
evaluatorType?: string;
evaluatorName?: string;
evaluatorKey?: string;
}[];
createdById?: string;
}[];
hasMore?: boolean;
}
Create an Experiment
Creates an experiment. project_id defaults to the API key’s project and is required for keys that span several projects. When dataset_id is set, a draft run is created with rows from that dataset.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.create(display_name="<value>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.create({
displayName: "<value>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"display_name": Optional[str],
"project_id": Optional[str],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"source_run": { # optional
"experiment_id": Optional[str],
"run_id": Optional[str],
},
}
{
displayName?: string;
projectId?: string;
datasetId?: string;
datasetVersionId?: string;
sourceRun?: {
experimentId?: string;
runId?: string;
};
}
Show Response
Show Response
{
"experiment_id": Optional[str],
"display_name": Optional[str],
"project_id": Optional[str],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"selected_run_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
"source": Optional[str],
"row_height_level": Optional[int],
"unique_evaluators": [{ # optional
"evaluator_id": Optional[str],
"evaluator_type": Optional[str],
"evaluator_name": Optional[str],
"evaluator_key": Optional[str],
}],
"created_by_id": Optional[str],
}
{
experimentId?: string;
displayName?: string;
projectId?: string;
datasetId?: string;
datasetVersionId?: string;
selectedRunId?: string;
createdAt?: Date;
updatedAt?: Date;
source?: string;
rowHeightLevel?: number;
uniqueEvaluators?: {
evaluatorId?: string;
evaluatorType?: string;
evaluatorName?: string;
evaluatorKey?: string;
}[];
createdById?: string;
}
Retrieve an Experiment
Returns experiment metadata. Runs and cell data are separate resources.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.get(experiment_id="<id>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.get({
experimentId: "<id>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
}
{
experimentId: string; // required
}
Show Response
Show Response
{
"experiment_id": Optional[str],
"display_name": Optional[str],
"project_id": Optional[str],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"selected_run_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
"source": Optional[str],
"row_height_level": Optional[int],
"unique_evaluators": [{ # optional
"evaluator_id": Optional[str],
"evaluator_type": Optional[str],
"evaluator_name": Optional[str],
"evaluator_key": Optional[str],
}],
"created_by_id": Optional[str],
}
{
experimentId?: string;
displayName?: string;
projectId?: string;
datasetId?: string;
datasetVersionId?: string;
selectedRunId?: string;
createdAt?: Date;
updatedAt?: Date;
source?: string;
rowHeightLevel?: number;
uniqueEvaluators?: {
evaluatorId?: string;
evaluatorType?: string;
evaluatorName?: string;
evaluatorKey?: string;
}[];
createdById?: string;
}
Delete an Experiment
Deletes an experiment and its runs.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
orq.experiments.delete(experiment_id="<id>")
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
await orq.experiments.delete({
experimentId: "<id>",
});
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
}
{
experimentId: string; // required
}
Update an Experiment
Updates experiment metadata. Omitted optional fields keep their current values.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.update(experiment_id="<id>", display_name="<value>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.update({
experimentId: "<id>",
updateExperimentRequest: {
displayName: "<value>",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"display_name": Optional[str],
"selected_run_id": Optional[str],
"row_height_level": Optional[int],
"project_id": Optional[str],
}
{
experimentId: string; // required
updateExperimentRequest: { // required
displayName?: string;
selectedRunId?: string;
rowHeightLevel?: number;
projectId?: string;
};
}
Show Response
Show Response
{
"experiment_id": Optional[str],
"display_name": Optional[str],
"project_id": Optional[str],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"selected_run_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
"source": Optional[str],
"row_height_level": Optional[int],
"unique_evaluators": [{ # optional
"evaluator_id": Optional[str],
"evaluator_type": Optional[str],
"evaluator_name": Optional[str],
"evaluator_key": Optional[str],
}],
"created_by_id": Optional[str],
}
{
experimentId?: string;
displayName?: string;
projectId?: string;
datasetId?: string;
datasetVersionId?: string;
selectedRunId?: string;
createdAt?: Date;
updatedAt?: Date;
source?: string;
rowHeightLevel?: number;
uniqueEvaluators?: {
evaluatorId?: string;
evaluatorType?: string;
evaluatorName?: string;
evaluatorKey?: string;
}[];
createdById?: string;
}
List Runs
Lists runs for an experiment, newest first. Each run is metadata only.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.list_runs(experiment_id="<id>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.listRuns({
experimentId: "<id>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"limit": Optional[int],
"starting_after": Optional[str],
"ending_before": Optional[str],
"include_metrics": Optional[bool],
}
{
experimentId: string; // required
limit?: number;
startingAfter?: string;
endingBefore?: string;
includeMetrics?: boolean;
}
Show Response
Show Response
{
"data": [{ # optional
"run_id": Optional[str],
"experiment_id": Optional[str],
"key": Optional[str],
"status": Optional[Literal["RUN_STATUS_UNSPECIFIED", "RUN_STATUS_DRAFT", "RUN_STATUS_QUEUED", "RUN_STATUS_RUNNING", "RUN_STATUS_COMPLETED", "RUN_STATUS_FAILED", "RUN_STATUS_CANCELED"]],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"row_count": Optional[int],
"cell_status_counts": { # optional
"not_run": Optional[int],
"queued": Optional[int],
"running": Optional[int],
"completed": Optional[int],
"failed": Optional[int],
"canceled": Optional[int],
},
"columns": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
"started_at": str, # optional
"completed_at": str, # optional
"created_at": str, # optional
"updated_at": str, # optional
"metrics": Dict[str, float], # optional
"stats": { # optional
"total_cost": Optional[float],
"total_runs": Optional[int],
"total_errors": Optional[int],
"total_done": Optional[int],
},
"created_by_id": Optional[str],
}],
"has_more": Optional[bool],
}
{
data?: {
runId?: string;
experimentId?: string;
key?: string;
status?: "RUN_STATUS_UNSPECIFIED" | "RUN_STATUS_DRAFT" | "RUN_STATUS_QUEUED" | "RUN_STATUS_RUNNING" | "RUN_STATUS_COMPLETED" | "RUN_STATUS_FAILED" | "RUN_STATUS_CANCELED";
datasetId?: string;
datasetVersionId?: string;
rowCount?: number;
cellStatusCounts?: {
notRun?: number;
queued?: number;
running?: number;
completed?: number;
failed?: number;
canceled?: number;
};
columns?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
startedAt?: Date;
completedAt?: Date;
createdAt?: Date;
updatedAt?: Date;
metrics?: Record<string, number>;
stats?: {
totalCost?: number;
totalRuns?: number;
totalErrors?: number;
totalDone?: number;
};
createdById?: string;
}[];
hasMore?: boolean;
}
Create Run
Creates a new run, copying columns from the selected run when present. A rerun is a new run.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.create_run(experiment_id="<id>", key="baseline")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.createRun({
experimentId: "<id>",
createRunRequest: {
key: "baseline",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"key": Optional[str],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
}
{
experimentId: string; // required
createRunRequest: { // required
key?: string;
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
};
}
Show Response
Show Response
{
"run_id": Optional[str],
"experiment_id": Optional[str],
"key": Optional[str],
"status": Optional[Literal["RUN_STATUS_UNSPECIFIED", "RUN_STATUS_DRAFT", "RUN_STATUS_QUEUED", "RUN_STATUS_RUNNING", "RUN_STATUS_COMPLETED", "RUN_STATUS_FAILED", "RUN_STATUS_CANCELED"]],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"row_count": Optional[int],
"cell_status_counts": { # optional
"not_run": Optional[int],
"queued": Optional[int],
"running": Optional[int],
"completed": Optional[int],
"failed": Optional[int],
"canceled": Optional[int],
},
"columns": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
"started_at": str, # optional
"completed_at": str, # optional
"created_at": str, # optional
"updated_at": str, # optional
"metrics": Dict[str, float], # optional
"stats": { # optional
"total_cost": Optional[float],
"total_runs": Optional[int],
"total_errors": Optional[int],
"total_done": Optional[int],
},
"created_by_id": Optional[str],
}
{
runId?: string;
experimentId?: string;
key?: string;
status?: "RUN_STATUS_UNSPECIFIED" | "RUN_STATUS_DRAFT" | "RUN_STATUS_QUEUED" | "RUN_STATUS_RUNNING" | "RUN_STATUS_COMPLETED" | "RUN_STATUS_FAILED" | "RUN_STATUS_CANCELED";
datasetId?: string;
datasetVersionId?: string;
rowCount?: number;
cellStatusCounts?: {
notRun?: number;
queued?: number;
running?: number;
completed?: number;
failed?: number;
canceled?: number;
};
columns?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
startedAt?: Date;
completedAt?: Date;
createdAt?: Date;
updatedAt?: Date;
metrics?: Record<string, number>;
stats?: {
totalCost?: number;
totalRuns?: number;
totalErrors?: number;
totalDone?: number;
};
createdById?: string;
}
Get Run
Returns run status, per-cell status counts, row count, and column configs. Does not return the cell matrix.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.get_run(experiment_id="<id>", run_id="<id>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.getRun({
experimentId: "<id>",
runId: "<id>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
}
{
experimentId: string; // required
runId: string; // required
}
Show Response
Show Response
{
"run_id": Optional[str],
"experiment_id": Optional[str],
"key": Optional[str],
"status": Optional[Literal["RUN_STATUS_UNSPECIFIED", "RUN_STATUS_DRAFT", "RUN_STATUS_QUEUED", "RUN_STATUS_RUNNING", "RUN_STATUS_COMPLETED", "RUN_STATUS_FAILED", "RUN_STATUS_CANCELED"]],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"row_count": Optional[int],
"cell_status_counts": { # optional
"not_run": Optional[int],
"queued": Optional[int],
"running": Optional[int],
"completed": Optional[int],
"failed": Optional[int],
"canceled": Optional[int],
},
"columns": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
"started_at": str, # optional
"completed_at": str, # optional
"created_at": str, # optional
"updated_at": str, # optional
"metrics": Dict[str, float], # optional
"stats": { # optional
"total_cost": Optional[float],
"total_runs": Optional[int],
"total_errors": Optional[int],
"total_done": Optional[int],
},
"created_by_id": Optional[str],
}
{
runId?: string;
experimentId?: string;
key?: string;
status?: "RUN_STATUS_UNSPECIFIED" | "RUN_STATUS_DRAFT" | "RUN_STATUS_QUEUED" | "RUN_STATUS_RUNNING" | "RUN_STATUS_COMPLETED" | "RUN_STATUS_FAILED" | "RUN_STATUS_CANCELED";
datasetId?: string;
datasetVersionId?: string;
rowCount?: number;
cellStatusCounts?: {
notRun?: number;
queued?: number;
running?: number;
completed?: number;
failed?: number;
canceled?: number;
};
columns?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
startedAt?: Date;
completedAt?: Date;
createdAt?: Date;
updatedAt?: Date;
metrics?: Record<string, number>;
stats?: {
totalCost?: number;
totalRuns?: number;
totalErrors?: number;
totalDone?: number;
};
createdById?: string;
}
Delete Run
Deletes a run and its cell data. Distinct from cancel, which keeps completed cells.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
orq.experiments.delete_run(experiment_id="<id>", run_id="<id>")
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
await orq.experiments.deleteRun({
experimentId: "<id>",
runId: "<id>",
});
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
}
{
experimentId: string; // required
runId: string; // required
}
Update Run
Updates run key and execution settings. Does not execute the run.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.update_run(experiment_id="<id>", run_id="<id>", key="baseline")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.updateRun({
experimentId: "<id>",
runId: "<id>",
updateRunRequest: {
key: "baseline",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"key": Optional[str],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
}
{
experimentId: string; // required
runId: string; // required
updateRunRequest: { // required
key?: string;
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
};
}
Show Response
Show Response
{
"run_id": Optional[str],
"experiment_id": Optional[str],
"key": Optional[str],
"status": Optional[Literal["RUN_STATUS_UNSPECIFIED", "RUN_STATUS_DRAFT", "RUN_STATUS_QUEUED", "RUN_STATUS_RUNNING", "RUN_STATUS_COMPLETED", "RUN_STATUS_FAILED", "RUN_STATUS_CANCELED"]],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"row_count": Optional[int],
"cell_status_counts": { # optional
"not_run": Optional[int],
"queued": Optional[int],
"running": Optional[int],
"completed": Optional[int],
"failed": Optional[int],
"canceled": Optional[int],
},
"columns": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
"started_at": str, # optional
"completed_at": str, # optional
"created_at": str, # optional
"updated_at": str, # optional
"metrics": Dict[str, float], # optional
"stats": { # optional
"total_cost": Optional[float],
"total_runs": Optional[int],
"total_errors": Optional[int],
"total_done": Optional[int],
},
"created_by_id": Optional[str],
}
{
runId?: string;
experimentId?: string;
key?: string;
status?: "RUN_STATUS_UNSPECIFIED" | "RUN_STATUS_DRAFT" | "RUN_STATUS_QUEUED" | "RUN_STATUS_RUNNING" | "RUN_STATUS_COMPLETED" | "RUN_STATUS_FAILED" | "RUN_STATUS_CANCELED";
datasetId?: string;
datasetVersionId?: string;
rowCount?: number;
cellStatusCounts?: {
notRun?: number;
queued?: number;
running?: number;
completed?: number;
failed?: number;
canceled?: number;
};
columns?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
startedAt?: Date;
completedAt?: Date;
createdAt?: Date;
updatedAt?: Date;
metrics?: Record<string, number>;
stats?: {
totalCost?: number;
totalRuns?: number;
totalErrors?: number;
totalDone?: number;
};
createdById?: string;
}
Cancel Run
Stops queued and running cells and moves the run to canceled. A cell whose execution is already in flight may still write its result.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.cancel_run(experiment_id="<id>", run_id="<id>", row_ids=["<row_id>"])
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.cancelRun({
experimentId: "<id>",
runId: "<id>",
cancelExperimentRunRequest: {
rowIds: ["<value>"],
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"row_ids": List[str], # optional
}
{
experimentId: string; // required
runId: string; // required
cancelExperimentRunRequest: { // required
rowIds?: string[];
};
}
Show Response
Show Response
{
"run_id": Optional[str],
"experiment_id": Optional[str],
"key": Optional[str],
"status": Optional[Literal["RUN_STATUS_UNSPECIFIED", "RUN_STATUS_DRAFT", "RUN_STATUS_QUEUED", "RUN_STATUS_RUNNING", "RUN_STATUS_COMPLETED", "RUN_STATUS_FAILED", "RUN_STATUS_CANCELED"]],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"row_count": Optional[int],
"cell_status_counts": { # optional
"not_run": Optional[int],
"queued": Optional[int],
"running": Optional[int],
"completed": Optional[int],
"failed": Optional[int],
"canceled": Optional[int],
},
"columns": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
"started_at": str, # optional
"completed_at": str, # optional
"created_at": str, # optional
"updated_at": str, # optional
"metrics": Dict[str, float], # optional
"stats": { # optional
"total_cost": Optional[float],
"total_runs": Optional[int],
"total_errors": Optional[int],
"total_done": Optional[int],
},
"created_by_id": Optional[str],
}
{
runId?: string;
experimentId?: string;
key?: string;
status?: "RUN_STATUS_UNSPECIFIED" | "RUN_STATUS_DRAFT" | "RUN_STATUS_QUEUED" | "RUN_STATUS_RUNNING" | "RUN_STATUS_COMPLETED" | "RUN_STATUS_FAILED" | "RUN_STATUS_CANCELED";
datasetId?: string;
datasetVersionId?: string;
rowCount?: number;
cellStatusCounts?: {
notRun?: number;
queued?: number;
running?: number;
completed?: number;
failed?: number;
canceled?: number;
};
columns?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
startedAt?: Date;
completedAt?: Date;
createdAt?: Date;
updatedAt?: Date;
metrics?: Record<string, number>;
stats?: {
totalCost?: number;
totalRuns?: number;
totalErrors?: number;
totalDone?: number;
};
createdById?: string;
}
Update Run Column Layout
Reorders, resizes, hides or shows columns of a run in one call. Columns not listed keep their layout.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.update_run_column_layout(experiment_id="<id>", run_id="<id>", columns=[{"column_id": "<id>", "position": 0}])
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.updateRunColumnLayout({
experimentId: "<id>",
runId: "<id>",
updateRunColumnLayoutRequest: {
columns: [{ columnId: "<id>", position: 0 }],
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"columns": [{ # optional
"column_id": Optional[str],
"position": Optional[int],
"width": Optional[int],
"active": Optional[bool],
}],
}
{
experimentId: string; // required
runId: string; // required
updateRunColumnLayoutRequest: { // required
columns?: {
columnId?: string;
position?: number;
width?: number;
active?: boolean;
}[];
};
}
Show Response
Show Response
{
"run_id": Optional[str],
"experiment_id": Optional[str],
"key": Optional[str],
"status": Optional[Literal["RUN_STATUS_UNSPECIFIED", "RUN_STATUS_DRAFT", "RUN_STATUS_QUEUED", "RUN_STATUS_RUNNING", "RUN_STATUS_COMPLETED", "RUN_STATUS_FAILED", "RUN_STATUS_CANCELED"]],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"row_count": Optional[int],
"cell_status_counts": { # optional
"not_run": Optional[int],
"queued": Optional[int],
"running": Optional[int],
"completed": Optional[int],
"failed": Optional[int],
"canceled": Optional[int],
},
"columns": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
"started_at": str, # optional
"completed_at": str, # optional
"created_at": str, # optional
"updated_at": str, # optional
"metrics": Dict[str, float], # optional
"stats": { # optional
"total_cost": Optional[float],
"total_runs": Optional[int],
"total_errors": Optional[int],
"total_done": Optional[int],
},
"created_by_id": Optional[str],
}
{
runId?: string;
experimentId?: string;
key?: string;
status?: "RUN_STATUS_UNSPECIFIED" | "RUN_STATUS_DRAFT" | "RUN_STATUS_QUEUED" | "RUN_STATUS_RUNNING" | "RUN_STATUS_COMPLETED" | "RUN_STATUS_FAILED" | "RUN_STATUS_CANCELED";
datasetId?: string;
datasetVersionId?: string;
rowCount?: number;
cellStatusCounts?: {
notRun?: number;
queued?: number;
running?: number;
completed?: number;
failed?: number;
canceled?: number;
};
columns?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
startedAt?: Date;
completedAt?: Date;
createdAt?: Date;
updatedAt?: Date;
metrics?: Record<string, number>;
stats?: {
totalCost?: number;
totalRuns?: number;
totalErrors?: number;
totalDone?: number;
};
createdById?: string;
}
List Evals
List eval columnsfrom orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.list_evals(experiment_id="<id>", run_id="<id>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.listEvals({
experimentId: "<id>",
runId: "<id>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
}
{
experimentId: string; // required
runId: string; // required
}
Show Response
Show Response
{
"data": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
}
{
data?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
}
Create Eval
Links an existing evaluator. Config is updated at the evaluator source, not here.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.create_eval(experiment_id="<id>", run_id="<id>", position=0)
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.createEval({
experimentId: "<id>",
runId: "<id>",
createEvalColumnRequest: {
position: 0,
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"evaluator_id": Optional[str],
"position": Optional[int],
}
{
experimentId: string; // required
runId: string; // required
createEvalColumnRequest: { // required
evaluatorId?: string;
position?: number;
};
}
Show Response
Show Response
{
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}
{
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}
Delete Eval
Remove an eval columnfrom orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
orq.experiments.delete_eval(experiment_id="<id>", run_id="<id>", eval_id="<id>")
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
await orq.experiments.deleteEval({
experimentId: "<id>",
runId: "<id>",
evalId: "<id>",
});
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"eval_id": str, # required
}
{
experimentId: string; // required
runId: string; // required
evalId: string; // required
}
Patch Eval Cell
Human override of an automated evaluator verdict on one datapoint.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.patch_eval_cell(experiment_id="<id>", run_id="<id>", eval_id="<id>", datapoint_id="<id>", value={"quality": "good"})
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.patchEvalCell({
experimentId: "<id>",
runId: "<id>",
evalId: "<id>",
datapointId: "<id>",
patchEvalCellRequest: {
value: { "<key>": "<value>" },
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"eval_id": str, # required
"datapoint_id": str, # required
"value": Dict[str, Any], # optional
}
{
experimentId: string; // required
runId: string; // required
evalId: string; // required
datapointId: string; // required
patchEvalCellRequest: { // required
value?: Record<string, any>;
};
}
Show Response
Show Response
{
"cell_id": Optional[str],
"row_id": Optional[str],
"column_id": Optional[str],
"repeat_index": Optional[int],
"status": Optional[Literal["CELL_STATUS_UNSPECIFIED", "CELL_STATUS_NOT_RUN", "CELL_STATUS_QUEUED", "CELL_STATUS_RUNNING", "CELL_STATUS_COMPLETED", "CELL_STATUS_FAILED", "CELL_STATUS_CANCELED"]],
"value": Dict[str, Any], # optional
"artifacts": { # optional
"trace_id": Optional[str],
"response_id": Optional[str],
"session_id": Optional[str],
"session_run_id": Optional[str],
"pull_request_url": Optional[str],
"patch_object_key": Optional[str],
},
"progress": { # optional
"elapsed": Optional[str],
"spend_usd": Optional[float],
"activity": Optional[str],
},
"error_message": Optional[str],
"type": Optional[Literal["CELL_TYPE_UNSPECIFIED", "CELL_TYPE_TEXT", "CELL_TYPE_NUMBER", "CELL_TYPE_INPUTS", "CELL_TYPE_MESSAGES", "CELL_TYPE_RESPONSE", "CELL_TYPE_AGENT_RESPONSE", "CELL_TYPE_EVALUATOR", "CELL_TYPE_METRIC", "CELL_TYPE_HUMAN_EVAL", "CELL_TYPE_OPENRESPONSE"]],
"task_column_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
}
{
cellId?: string;
rowId?: string;
columnId?: string;
repeatIndex?: number;
status?: "CELL_STATUS_UNSPECIFIED" | "CELL_STATUS_NOT_RUN" | "CELL_STATUS_QUEUED" | "CELL_STATUS_RUNNING" | "CELL_STATUS_COMPLETED" | "CELL_STATUS_FAILED" | "CELL_STATUS_CANCELED";
value?: Record<string, any>;
artifacts?: {
traceId?: string;
responseId?: string;
sessionId?: string;
sessionRunId?: string;
pullRequestUrl?: string;
patchObjectKey?: string;
};
progress?: {
elapsed?: string;
spendUsd?: number;
activity?: string;
};
errorMessage?: string;
type?: "CELL_TYPE_UNSPECIFIED" | "CELL_TYPE_TEXT" | "CELL_TYPE_NUMBER" | "CELL_TYPE_INPUTS" | "CELL_TYPE_MESSAGES" | "CELL_TYPE_RESPONSE" | "CELL_TYPE_AGENT_RESPONSE" | "CELL_TYPE_EVALUATOR" | "CELL_TYPE_METRIC" | "CELL_TYPE_HUMAN_EVAL" | "CELL_TYPE_OPENRESPONSE";
taskColumnId?: string;
createdAt?: Date;
updatedAt?: Date;
}
Execute Run
Executes cells at the requested scope. FULL, PARTIAL, ROW and COLUMN queue the matching cells (a task column also queues its linked evaluator, metric and review cells) and start the orchestrator, which also re-runs failed cells of every row it picks up. A run that is already queued or running rejects new execution until it finishes or is cancelled. CELL re-runs one task or evaluator cell in place. Re-executing a terminal run does not require creating a new run.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.execute_run(experiment_id="<id>", run_id="<id>", scope="EXECUTE_SCOPE_CELL")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.executeRun({
experimentId: "<id>",
runId: "<id>",
executeRunRequest: {
scope: "EXECUTE_SCOPE_CELL",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"scope": Optional[Literal["EXECUTE_SCOPE_UNSPECIFIED", "EXECUTE_SCOPE_CELL", "EXECUTE_SCOPE_ROW", "EXECUTE_SCOPE_COLUMN", "EXECUTE_SCOPE_PARTIAL", "EXECUTE_SCOPE_FULL"]],
"targets": [{ # optional
"row_id": Optional[str],
"column_id": Optional[str],
"column_key": Optional[str],
"repeat_index": Optional[int],
}],
"constraints": { # optional
"llm": { # optional
"max_turns": Optional[int],
"timeout": Optional[str],
"concurrency": Optional[int],
},
"harness": { # optional
"wall_clock_value": Optional[int],
"wall_clock_unit": Optional[Literal["WALL_CLOCK_UNIT_UNSPECIFIED", "WALL_CLOCK_UNIT_MINUTES", "WALL_CLOCK_UNIT_HOURS"]],
"spend_cap_usd": Optional[float],
"spend_cap_tokens": Optional[str],
"concurrency": Optional[int],
},
},
"repeat_count": Optional[int],
}
{
experimentId: string; // required
runId: string; // required
executeRunRequest: { // required
scope?: "EXECUTE_SCOPE_UNSPECIFIED" | "EXECUTE_SCOPE_CELL" | "EXECUTE_SCOPE_ROW" | "EXECUTE_SCOPE_COLUMN" | "EXECUTE_SCOPE_PARTIAL" | "EXECUTE_SCOPE_FULL";
targets?: {
rowId?: string;
columnId?: string;
columnKey?: string;
repeatIndex?: number;
}[];
constraints?: {
llm?: {
maxTurns?: number;
timeout?: string;
concurrency?: number;
};
harness?: {
wallClockValue?: number;
wallClockUnit?: "WALL_CLOCK_UNIT_UNSPECIFIED" | "WALL_CLOCK_UNIT_MINUTES" | "WALL_CLOCK_UNIT_HOURS";
spendCapUsd?: number;
spendCapTokens?: string;
concurrency?: number;
};
};
repeatCount?: number;
};
}
Show Response
Show Response
{
"run_id": Optional[str],
"experiment_id": Optional[str],
"key": Optional[str],
"status": Optional[Literal["RUN_STATUS_UNSPECIFIED", "RUN_STATUS_DRAFT", "RUN_STATUS_QUEUED", "RUN_STATUS_RUNNING", "RUN_STATUS_COMPLETED", "RUN_STATUS_FAILED", "RUN_STATUS_CANCELED"]],
"dataset_id": Optional[str],
"dataset_version_id": Optional[str],
"row_count": Optional[int],
"cell_status_counts": { # optional
"not_run": Optional[int],
"queued": Optional[int],
"running": Optional[int],
"completed": Optional[int],
"failed": Optional[int],
"canceled": Optional[int],
},
"columns": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
"settings": { # optional
"concurrency": Optional[int],
"timeout": Optional[str],
"max_turns": Optional[int],
"hidden_column_ids": List[str], # optional
},
"started_at": str, # optional
"completed_at": str, # optional
"created_at": str, # optional
"updated_at": str, # optional
"metrics": Dict[str, float], # optional
"stats": { # optional
"total_cost": Optional[float],
"total_runs": Optional[int],
"total_errors": Optional[int],
"total_done": Optional[int],
},
"created_by_id": Optional[str],
}
{
runId?: string;
experimentId?: string;
key?: string;
status?: "RUN_STATUS_UNSPECIFIED" | "RUN_STATUS_DRAFT" | "RUN_STATUS_QUEUED" | "RUN_STATUS_RUNNING" | "RUN_STATUS_COMPLETED" | "RUN_STATUS_FAILED" | "RUN_STATUS_CANCELED";
datasetId?: string;
datasetVersionId?: string;
rowCount?: number;
cellStatusCounts?: {
notRun?: number;
queued?: number;
running?: number;
completed?: number;
failed?: number;
canceled?: number;
};
columns?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
settings?: {
concurrency?: number;
timeout?: string;
maxTurns?: number;
hiddenColumnIds?: string[];
};
startedAt?: Date;
completedAt?: Date;
createdAt?: Date;
updatedAt?: Date;
metrics?: Record<string, number>;
stats?: {
totalCost?: number;
totalRuns?: number;
totalErrors?: number;
totalDone?: number;
};
createdById?: string;
}
Get Run Metrics
Returns aggregated evaluator and built-in metrics for a run, keyed by task column and metric. Pass baseline_run_id to include the delta against another run.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.get_run_metrics(experiment_id="<id>", run_id="<id>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.getRunMetrics({
experimentId: "<id>",
runId: "<id>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"baseline_run_id": Optional[str],
}
{
experimentId: string; // required
runId: string; // required
baselineRunId?: string;
}
Show Response
Show Response
{
"metrics": Dict[str, MetricSummary], # optional
}
{
metrics?: Record<string, MetricSummary>;
}
Query Run
Returns selected cells by row and column name. POST because selections can exceed URL limits.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.query_run(experiment_id="<id>", run_id="<id>", rows=["<row_id>"])
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.queryRun({
experimentId: "<id>",
runId: "<id>",
queryRunRequest: {
rows: ["<value>"],
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"rows": List[str], # optional
"columns": List[str], # optional
"repeat_index": Optional[int],
"limit": Optional[int],
"starting_after": Optional[str],
"updated_after": str, # optional
"partial_rows": Optional[bool],
"exclude_static_cells": Optional[bool],
}
{
experimentId: string; // required
runId: string; // required
queryRunRequest: { // required
rows?: string[];
columns?: string[];
repeatIndex?: number;
limit?: number;
startingAfter?: string;
updatedAfter?: Date;
partialRows?: boolean;
excludeStaticCells?: boolean;
};
}
Show Response
Show Response
{
"rows": [{ # optional
"row_id": Optional[str],
"display_name": Optional[str],
"cells": [{ # optional
"cell_id": Optional[str],
"row_id": Optional[str],
"column_id": Optional[str],
"repeat_index": Optional[int],
"status": Optional[Literal["CELL_STATUS_UNSPECIFIED", "CELL_STATUS_NOT_RUN", "CELL_STATUS_QUEUED", "CELL_STATUS_RUNNING", "CELL_STATUS_COMPLETED", "CELL_STATUS_FAILED", "CELL_STATUS_CANCELED"]],
"value": Dict[str, Any], # optional
"artifacts": { # optional
"trace_id": Optional[str],
"response_id": Optional[str],
"session_id": Optional[str],
"session_run_id": Optional[str],
"pull_request_url": Optional[str],
"patch_object_key": Optional[str],
},
"progress": { # optional
"elapsed": Optional[str],
"spend_usd": Optional[float],
"activity": Optional[str],
},
"error_message": Optional[str],
"type": Optional[Literal["CELL_TYPE_UNSPECIFIED", "CELL_TYPE_TEXT", "CELL_TYPE_NUMBER", "CELL_TYPE_INPUTS", "CELL_TYPE_MESSAGES", "CELL_TYPE_RESPONSE", "CELL_TYPE_AGENT_RESPONSE", "CELL_TYPE_EVALUATOR", "CELL_TYPE_METRIC", "CELL_TYPE_HUMAN_EVAL", "CELL_TYPE_OPENRESPONSE"]],
"task_column_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
}],
"status": Optional[Literal["ROW_STATUS_UNSPECIFIED", "ROW_STATUS_NOT_RUN", "ROW_STATUS_QUEUED", "ROW_STATUS_RUNNING", "ROW_STATUS_COMPLETED", "ROW_STATUS_FAILED", "ROW_STATUS_CANCELED"]],
"created_at": str, # optional
"updated_at": str, # optional
}],
"has_more": Optional[bool],
"server_timestamp": str, # optional
}
{
rows?: {
rowId?: string;
displayName?: string;
cells?: {
cellId?: string;
rowId?: string;
columnId?: string;
repeatIndex?: number;
status?: "CELL_STATUS_UNSPECIFIED" | "CELL_STATUS_NOT_RUN" | "CELL_STATUS_QUEUED" | "CELL_STATUS_RUNNING" | "CELL_STATUS_COMPLETED" | "CELL_STATUS_FAILED" | "CELL_STATUS_CANCELED";
value?: Record<string, any>;
artifacts?: {
traceId?: string;
responseId?: string;
sessionId?: string;
sessionRunId?: string;
pullRequestUrl?: string;
patchObjectKey?: string;
};
progress?: {
elapsed?: string;
spendUsd?: number;
activity?: string;
};
errorMessage?: string;
type?: "CELL_TYPE_UNSPECIFIED" | "CELL_TYPE_TEXT" | "CELL_TYPE_NUMBER" | "CELL_TYPE_INPUTS" | "CELL_TYPE_MESSAGES" | "CELL_TYPE_RESPONSE" | "CELL_TYPE_AGENT_RESPONSE" | "CELL_TYPE_EVALUATOR" | "CELL_TYPE_METRIC" | "CELL_TYPE_HUMAN_EVAL" | "CELL_TYPE_OPENRESPONSE";
taskColumnId?: string;
createdAt?: Date;
updatedAt?: Date;
}[];
status?: "ROW_STATUS_UNSPECIFIED" | "ROW_STATUS_NOT_RUN" | "ROW_STATUS_QUEUED" | "ROW_STATUS_RUNNING" | "ROW_STATUS_COMPLETED" | "ROW_STATUS_FAILED" | "ROW_STATUS_CANCELED";
createdAt?: Date;
updatedAt?: Date;
}[];
hasMore?: boolean;
serverTimestamp?: Date;
}
List Reviews
List human review columnsfrom orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.list_reviews(experiment_id="<id>", run_id="<id>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.listReviews({
experimentId: "<id>",
runId: "<id>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
}
{
experimentId: string; // required
runId: string; // required
}
Show Response
Show Response
{
"data": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
}
{
data?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
}
Create Review
Add a human review columnfrom orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.create_review(experiment_id="<id>", run_id="<id>", display_name="<value>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.createReview({
experimentId: "<id>",
runId: "<id>",
createReviewColumnRequest: {
displayName: "<value>",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"human_eval_id": Optional[str],
"display_name": Optional[str],
"human_eval_type": Optional[str],
"position": Optional[int],
}
{
experimentId: string; // required
runId: string; // required
createReviewColumnRequest: { // required
humanEvalId?: string;
displayName?: string;
humanEvalType?: string;
position?: number;
};
}
Show Response
Show Response
{
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}
{
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}
Delete Review
Remove a human review columnfrom orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
orq.experiments.delete_review(experiment_id="<id>", run_id="<id>", review_id="<id>")
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
await orq.experiments.deleteReview({
experimentId: "<id>",
runId: "<id>",
reviewId: "<id>",
});
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"review_id": str, # required
}
{
experimentId: string; // required
runId: string; // required
reviewId: string; // required
}
Update Review
Updates display fields on the column. The human-eval definition is updated at source.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.update_review(experiment_id="<id>", run_id="<id>", review_id="<id>", display_name="<value>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.updateReview({
experimentId: "<id>",
runId: "<id>",
reviewId: "<id>",
updateReviewColumnRequest: {
displayName: "<value>",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"review_id": str, # required
"display_name": Optional[str],
}
{
experimentId: string; // required
runId: string; // required
reviewId: string; // required
updateReviewColumnRequest: { // required
displayName?: string;
};
}
Show Response
Show Response
{
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}
{
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}
Patch Review Cell
Set a human review cellfrom orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.patch_review_cell(experiment_id="<id>", run_id="<id>", review_id="<id>", datapoint_id="<id>", value={"quality": "good"})
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.patchReviewCell({
experimentId: "<id>",
runId: "<id>",
reviewId: "<id>",
datapointId: "<id>",
patchReviewCellRequest: {
value: { "<key>": "<value>" },
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"review_id": str, # required
"datapoint_id": str, # required
"value": Dict[str, Any], # optional
}
{
experimentId: string; // required
runId: string; // required
reviewId: string; // required
datapointId: string; // required
patchReviewCellRequest: { // required
value?: Record<string, any>;
};
}
Show Response
Show Response
{
"cell_id": Optional[str],
"row_id": Optional[str],
"column_id": Optional[str],
"repeat_index": Optional[int],
"status": Optional[Literal["CELL_STATUS_UNSPECIFIED", "CELL_STATUS_NOT_RUN", "CELL_STATUS_QUEUED", "CELL_STATUS_RUNNING", "CELL_STATUS_COMPLETED", "CELL_STATUS_FAILED", "CELL_STATUS_CANCELED"]],
"value": Dict[str, Any], # optional
"artifacts": { # optional
"trace_id": Optional[str],
"response_id": Optional[str],
"session_id": Optional[str],
"session_run_id": Optional[str],
"pull_request_url": Optional[str],
"patch_object_key": Optional[str],
},
"progress": { # optional
"elapsed": Optional[str],
"spend_usd": Optional[float],
"activity": Optional[str],
},
"error_message": Optional[str],
"type": Optional[Literal["CELL_TYPE_UNSPECIFIED", "CELL_TYPE_TEXT", "CELL_TYPE_NUMBER", "CELL_TYPE_INPUTS", "CELL_TYPE_MESSAGES", "CELL_TYPE_RESPONSE", "CELL_TYPE_AGENT_RESPONSE", "CELL_TYPE_EVALUATOR", "CELL_TYPE_METRIC", "CELL_TYPE_HUMAN_EVAL", "CELL_TYPE_OPENRESPONSE"]],
"task_column_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
}
{
cellId?: string;
rowId?: string;
columnId?: string;
repeatIndex?: number;
status?: "CELL_STATUS_UNSPECIFIED" | "CELL_STATUS_NOT_RUN" | "CELL_STATUS_QUEUED" | "CELL_STATUS_RUNNING" | "CELL_STATUS_COMPLETED" | "CELL_STATUS_FAILED" | "CELL_STATUS_CANCELED";
value?: Record<string, any>;
artifacts?: {
traceId?: string;
responseId?: string;
sessionId?: string;
sessionRunId?: string;
pullRequestUrl?: string;
patchObjectKey?: string;
};
progress?: {
elapsed?: string;
spendUsd?: number;
activity?: string;
};
errorMessage?: string;
type?: "CELL_TYPE_UNSPECIFIED" | "CELL_TYPE_TEXT" | "CELL_TYPE_NUMBER" | "CELL_TYPE_INPUTS" | "CELL_TYPE_MESSAGES" | "CELL_TYPE_RESPONSE" | "CELL_TYPE_AGENT_RESPONSE" | "CELL_TYPE_EVALUATOR" | "CELL_TYPE_METRIC" | "CELL_TYPE_HUMAN_EVAL" | "CELL_TYPE_OPENRESPONSE";
taskColumnId?: string;
createdAt?: Date;
updatedAt?: Date;
}
Delete Rows
Deletes the given rows of a run and their cells.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
orq.experiments.delete_rows(experiment_id="<id>", run_id="<id>", row_ids=["<row_id>"])
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
await orq.experiments.deleteRows({
experimentId: "<id>",
runId: "<id>",
deleteRowsRequest: {
rowIds: ["<value>"],
},
});
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"row_ids": List[str], # optional
}
{
experimentId: string; // required
runId: string; // required
deleteRowsRequest: { // required
rowIds?: string[];
};
}
Update Run Cell
Writes a cell value by cell id, including dataset and review cells. An empty value clears the cell. Running cells cannot be changed.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.update_run_cell(experiment_id="<id>", run_id="<id>", row_id="<id>", cell_id="<id>", value={"<key>": "<value>"})
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.updateRunCell({
experimentId: "<id>",
runId: "<id>",
rowId: "<id>",
cellId: "<id>",
updateRunCellRequest: {
value: { "<key>": "<value>" },
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"row_id": str, # required
"cell_id": str, # required
"value": Dict[str, Any], # optional
}
{
experimentId: string; // required
runId: string; // required
rowId: string; // required
cellId: string; // required
updateRunCellRequest: { // required
value?: Record<string, any>;
};
}
Show Response
Show Response
{
"cell_id": Optional[str],
"row_id": Optional[str],
"column_id": Optional[str],
"repeat_index": Optional[int],
"status": Optional[Literal["CELL_STATUS_UNSPECIFIED", "CELL_STATUS_NOT_RUN", "CELL_STATUS_QUEUED", "CELL_STATUS_RUNNING", "CELL_STATUS_COMPLETED", "CELL_STATUS_FAILED", "CELL_STATUS_CANCELED"]],
"value": Dict[str, Any], # optional
"artifacts": { # optional
"trace_id": Optional[str],
"response_id": Optional[str],
"session_id": Optional[str],
"session_run_id": Optional[str],
"pull_request_url": Optional[str],
"patch_object_key": Optional[str],
},
"progress": { # optional
"elapsed": Optional[str],
"spend_usd": Optional[float],
"activity": Optional[str],
},
"error_message": Optional[str],
"type": Optional[Literal["CELL_TYPE_UNSPECIFIED", "CELL_TYPE_TEXT", "CELL_TYPE_NUMBER", "CELL_TYPE_INPUTS", "CELL_TYPE_MESSAGES", "CELL_TYPE_RESPONSE", "CELL_TYPE_AGENT_RESPONSE", "CELL_TYPE_EVALUATOR", "CELL_TYPE_METRIC", "CELL_TYPE_HUMAN_EVAL", "CELL_TYPE_OPENRESPONSE"]],
"task_column_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
}
{
cellId?: string;
rowId?: string;
columnId?: string;
repeatIndex?: number;
status?: "CELL_STATUS_UNSPECIFIED" | "CELL_STATUS_NOT_RUN" | "CELL_STATUS_QUEUED" | "CELL_STATUS_RUNNING" | "CELL_STATUS_COMPLETED" | "CELL_STATUS_FAILED" | "CELL_STATUS_CANCELED";
value?: Record<string, any>;
artifacts?: {
traceId?: string;
responseId?: string;
sessionId?: string;
sessionRunId?: string;
pullRequestUrl?: string;
patchObjectKey?: string;
};
progress?: {
elapsed?: string;
spendUsd?: number;
activity?: string;
};
errorMessage?: string;
type?: "CELL_TYPE_UNSPECIFIED" | "CELL_TYPE_TEXT" | "CELL_TYPE_NUMBER" | "CELL_TYPE_INPUTS" | "CELL_TYPE_MESSAGES" | "CELL_TYPE_RESPONSE" | "CELL_TYPE_AGENT_RESPONSE" | "CELL_TYPE_EVALUATOR" | "CELL_TYPE_METRIC" | "CELL_TYPE_HUMAN_EVAL" | "CELL_TYPE_OPENRESPONSE";
taskColumnId?: string;
createdAt?: Date;
updatedAt?: Date;
}
Duplicate Row
Copies a row’s dataset cells into a new row placed after the source. Task, evaluator, metric and review cells start empty.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.duplicate_row(experiment_id="<id>", run_id="<id>", row_id="<id>", duplicate_row_request={"<key>": "<value>"})
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.duplicateRow({
experimentId: "<id>",
runId: "<id>",
rowId: "<id>",
duplicateRowRequest: {
"<key>": "<value>",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"row_id": str, # required
"duplicate_row_request": Dict[str, Any], # required
}
{
experimentId: string; // required
runId: string; // required
rowId: string; // required
duplicateRowRequest: Record<string, unknown>; // required
}
Show Response
Show Response
{
"row_id": Optional[str],
"display_name": Optional[str],
"cells": [{ # optional
"cell_id": Optional[str],
"row_id": Optional[str],
"column_id": Optional[str],
"repeat_index": Optional[int],
"status": Optional[Literal["CELL_STATUS_UNSPECIFIED", "CELL_STATUS_NOT_RUN", "CELL_STATUS_QUEUED", "CELL_STATUS_RUNNING", "CELL_STATUS_COMPLETED", "CELL_STATUS_FAILED", "CELL_STATUS_CANCELED"]],
"value": Dict[str, Any], # optional
"artifacts": { # optional
"trace_id": Optional[str],
"response_id": Optional[str],
"session_id": Optional[str],
"session_run_id": Optional[str],
"pull_request_url": Optional[str],
"patch_object_key": Optional[str],
},
"progress": { # optional
"elapsed": Optional[str],
"spend_usd": Optional[float],
"activity": Optional[str],
},
"error_message": Optional[str],
"type": Optional[Literal["CELL_TYPE_UNSPECIFIED", "CELL_TYPE_TEXT", "CELL_TYPE_NUMBER", "CELL_TYPE_INPUTS", "CELL_TYPE_MESSAGES", "CELL_TYPE_RESPONSE", "CELL_TYPE_AGENT_RESPONSE", "CELL_TYPE_EVALUATOR", "CELL_TYPE_METRIC", "CELL_TYPE_HUMAN_EVAL", "CELL_TYPE_OPENRESPONSE"]],
"task_column_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
}],
"status": Optional[Literal["ROW_STATUS_UNSPECIFIED", "ROW_STATUS_NOT_RUN", "ROW_STATUS_QUEUED", "ROW_STATUS_RUNNING", "ROW_STATUS_COMPLETED", "ROW_STATUS_FAILED", "ROW_STATUS_CANCELED"]],
"created_at": str, # optional
"updated_at": str, # optional
}
{
rowId?: string;
displayName?: string;
cells?: {
cellId?: string;
rowId?: string;
columnId?: string;
repeatIndex?: number;
status?: "CELL_STATUS_UNSPECIFIED" | "CELL_STATUS_NOT_RUN" | "CELL_STATUS_QUEUED" | "CELL_STATUS_RUNNING" | "CELL_STATUS_COMPLETED" | "CELL_STATUS_FAILED" | "CELL_STATUS_CANCELED";
value?: Record<string, any>;
artifacts?: {
traceId?: string;
responseId?: string;
sessionId?: string;
sessionRunId?: string;
pullRequestUrl?: string;
patchObjectKey?: string;
};
progress?: {
elapsed?: string;
spendUsd?: number;
activity?: string;
};
errorMessage?: string;
type?: "CELL_TYPE_UNSPECIFIED" | "CELL_TYPE_TEXT" | "CELL_TYPE_NUMBER" | "CELL_TYPE_INPUTS" | "CELL_TYPE_MESSAGES" | "CELL_TYPE_RESPONSE" | "CELL_TYPE_AGENT_RESPONSE" | "CELL_TYPE_EVALUATOR" | "CELL_TYPE_METRIC" | "CELL_TYPE_HUMAN_EVAL" | "CELL_TYPE_OPENRESPONSE";
taskColumnId?: string;
createdAt?: Date;
updatedAt?: Date;
}[];
status?: "ROW_STATUS_UNSPECIFIED" | "ROW_STATUS_NOT_RUN" | "ROW_STATUS_QUEUED" | "ROW_STATUS_RUNNING" | "ROW_STATUS_COMPLETED" | "ROW_STATUS_FAILED" | "ROW_STATUS_CANCELED";
createdAt?: Date;
updatedAt?: Date;
}
List Tasks
List task columnsfrom orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.list_tasks(experiment_id="<id>", run_id="<id>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.listTasks({
experimentId: "<id>",
runId: "<id>",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
}
{
experimentId: string; // required
runId: string; // required
}
Show Response
Show Response
{
"data": [{ # optional
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}],
}
{
data?: {
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}[];
}
Create Task
Attaches a prompt, orq agent, or (reserved) harness config as a task column. Input mapping is required.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.create_task(experiment_id="<id>", run_id="<id>", display_name="<value>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.createTask({
experimentId: "<id>",
runId: "<id>",
createTaskColumnRequest: {
displayName: "<value>",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"display_name": Optional[str],
"prompt": { # optional
"prompt_config": Dict[str, Any], # required
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"position": Optional[int],
}
{
experimentId: string; // required
runId: string; // required
createTaskColumnRequest: { // required
displayName?: string;
prompt?: {
promptConfig: Record<string, any>; // required
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
position?: number;
};
}
Show Response
Show Response
{
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}
{
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}
Delete Task
Delete a task columnfrom orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
orq.experiments.delete_task(experiment_id="<id>", run_id="<id>", task_id="<id>")
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
await orq.experiments.deleteTask({
experimentId: "<id>",
runId: "<id>",
taskId: "<id>",
});
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"task_id": str, # required
}
{
experimentId: string; // required
runId: string; // required
taskId: string; // required
}
Update Task
Updates an unlinked task column in place.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.update_task(experiment_id="<id>", run_id="<id>", task_id="<id>", display_name="<value>")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.updateTask({
experimentId: "<id>",
runId: "<id>",
taskId: "<id>",
updateTaskColumnRequest: {
displayName: "<value>",
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"task_id": str, # required
"display_name": Optional[str],
"prompt": { # optional
"prompt_config": Dict[str, Any], # required
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
}
{
experimentId: string; // required
runId: string; // required
taskId: string; // required
updateTaskColumnRequest: { // required
displayName?: string;
prompt?: {
promptConfig: Record<string, any>; // required
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
};
}
Show Response
Show Response
{
"column_id": Optional[str],
"key": Optional[str],
"display_name": Optional[str],
"position": Optional[int],
"active": Optional[bool],
"prompt": { # optional
"prompt_config": Dict[str, Any],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"engine": Optional[str],
"tools": List[Dict[str, Any]], # optional
"knowledge_bases": List[Dict[str, Any]], # optional
"metadata": Dict[str, Any], # optional
"entity_configuration": Dict[str, Any], # optional
},
"agent": { # optional
"agent_key": Optional[str],
"messages_source": Optional[Literal["MESSAGES_SOURCE_UNSPECIFIED", "MESSAGES_SOURCE_PROMPT_CONFIG", "MESSAGES_SOURCE_MESSAGES_COLUMN", "MESSAGES_SOURCE_COMBINED"]],
"messages": List[Dict[str, Any]], # optional
"engine": Optional[str],
"parent_id": Optional[str],
"agent_tag": Optional[str],
},
"harness": { # optional
"harness_config_id": Optional[str],
},
"openresponse": Dict[str, Any], # optional
"evaluator": { # optional
"evaluator_id": Optional[str],
"evaluator_name": Optional[str],
"evaluator_type": Optional[str],
"evaluator_output_type": Optional[str],
},
"review": { # optional
"human_eval_id": Optional[str],
"human_eval_name": Optional[str],
"human_eval_type": Optional[str],
},
"dataset": { # optional
"column_type": Optional[str],
},
"metric": { # optional
"metric_key": Optional[str],
},
"job_output": { # optional
"job_name": Optional[str],
},
"width": Optional[int],
}
{
columnId?: string;
key?: string;
displayName?: string;
position?: number;
active?: boolean;
prompt?: {
promptConfig: Record<string, any>;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
engine?: string;
tools?: Record<string, any>[];
knowledgeBases?: Record<string, any>[];
metadata?: Record<string, any>;
entityConfiguration?: Record<string, any>;
};
agent?: {
agentKey?: string;
messagesSource?: "MESSAGES_SOURCE_UNSPECIFIED" | "MESSAGES_SOURCE_PROMPT_CONFIG" | "MESSAGES_SOURCE_MESSAGES_COLUMN" | "MESSAGES_SOURCE_COMBINED";
messages?: Record<string, any>[];
engine?: string;
parentId?: string;
agentTag?: string;
};
harness?: {
harnessConfigId?: string;
};
openresponse?: Record<string, unknown>;
evaluator?: {
evaluatorId?: string;
evaluatorName?: string;
evaluatorType?: string;
evaluatorOutputType?: string;
};
review?: {
humanEvalId?: string;
humanEvalName?: string;
humanEvalType?: string;
};
dataset?: {
columnType?: string;
};
metric?: {
metricKey?: string;
};
jobOutput?: {
jobName?: string;
};
width?: number;
}
Patch Task Cell
Writes a cell result, including externally executed or harness cells.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.patch_task_cell(experiment_id="<id>", run_id="<id>", task_id="<id>", datapoint_id="<id>", repeat_index=0)
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.patchTaskCell({
experimentId: "<id>",
runId: "<id>",
taskId: "<id>",
datapointId: "<id>",
patchTaskCellRequest: {
repeatIndex: 0,
},
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"task_id": str, # required
"datapoint_id": str, # required
"repeat_index": Optional[int],
"value": Dict[str, Any], # optional
"artifacts": { # optional
"trace_id": Optional[str],
"response_id": Optional[str],
"session_id": Optional[str],
"session_run_id": Optional[str],
"pull_request_url": Optional[str],
"patch_object_key": Optional[str],
},
}
{
experimentId: string; // required
runId: string; // required
taskId: string; // required
datapointId: string; // required
patchTaskCellRequest: { // required
repeatIndex?: number;
value?: Record<string, any>;
artifacts?: {
traceId?: string;
responseId?: string;
sessionId?: string;
sessionRunId?: string;
pullRequestUrl?: string;
patchObjectKey?: string;
};
};
}
Show Response
Show Response
{
"cell_id": Optional[str],
"row_id": Optional[str],
"column_id": Optional[str],
"repeat_index": Optional[int],
"status": Optional[Literal["CELL_STATUS_UNSPECIFIED", "CELL_STATUS_NOT_RUN", "CELL_STATUS_QUEUED", "CELL_STATUS_RUNNING", "CELL_STATUS_COMPLETED", "CELL_STATUS_FAILED", "CELL_STATUS_CANCELED"]],
"value": Dict[str, Any], # optional
"artifacts": { # optional
"trace_id": Optional[str],
"response_id": Optional[str],
"session_id": Optional[str],
"session_run_id": Optional[str],
"pull_request_url": Optional[str],
"patch_object_key": Optional[str],
},
"progress": { # optional
"elapsed": Optional[str],
"spend_usd": Optional[float],
"activity": Optional[str],
},
"error_message": Optional[str],
"type": Optional[Literal["CELL_TYPE_UNSPECIFIED", "CELL_TYPE_TEXT", "CELL_TYPE_NUMBER", "CELL_TYPE_INPUTS", "CELL_TYPE_MESSAGES", "CELL_TYPE_RESPONSE", "CELL_TYPE_AGENT_RESPONSE", "CELL_TYPE_EVALUATOR", "CELL_TYPE_METRIC", "CELL_TYPE_HUMAN_EVAL", "CELL_TYPE_OPENRESPONSE"]],
"task_column_id": Optional[str],
"created_at": str, # optional
"updated_at": str, # optional
}
{
cellId?: string;
rowId?: string;
columnId?: string;
repeatIndex?: number;
status?: "CELL_STATUS_UNSPECIFIED" | "CELL_STATUS_NOT_RUN" | "CELL_STATUS_QUEUED" | "CELL_STATUS_RUNNING" | "CELL_STATUS_COMPLETED" | "CELL_STATUS_FAILED" | "CELL_STATUS_CANCELED";
value?: Record<string, any>;
artifacts?: {
traceId?: string;
responseId?: string;
sessionId?: string;
sessionRunId?: string;
pullRequestUrl?: string;
patchObjectKey?: string;
};
progress?: {
elapsed?: string;
spendUsd?: number;
activity?: string;
};
errorMessage?: string;
type?: "CELL_TYPE_UNSPECIFIED" | "CELL_TYPE_TEXT" | "CELL_TYPE_NUMBER" | "CELL_TYPE_INPUTS" | "CELL_TYPE_MESSAGES" | "CELL_TYPE_RESPONSE" | "CELL_TYPE_AGENT_RESPONSE" | "CELL_TYPE_EVALUATOR" | "CELL_TYPE_METRIC" | "CELL_TYPE_HUMAN_EVAL" | "CELL_TYPE_OPENRESPONSE";
taskColumnId?: string;
createdAt?: Date;
updatedAt?: Date;
}
Export Run
Downloads the run as a file, including human review and evaluator corrections made after an earlier export.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY"),
) as orq:
res = orq.experiments.export_run(experiment_id="<id>", run_id="<id>", format_="json")
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.experiments.exportRun({
experimentId: "<id>",
runId: "<id>",
format: "json",
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
"experiment_id": str, # required
"run_id": str, # required
"format_": Optional[Literal["csv", "json", "jsonl"]],
}
{
experimentId: string; // required
runId: string; // required
format?: "csv" | "json" | "jsonl";
}