Executions

The Executions endpoints let you track the status of optimization runs and retrieve results. Every model execution and async solve creates an execution record that you can query.

GET /api/v2/models/executions/all

List all executions across all models for your organization.

Authentication: Requires API key or JWT token.

Query Parameters

Response

Status values:

Examples

import httpx

API_URL = "https://jaot.io/api/v2"
headers = {"Authorization": "Bearer ok_live_your_key_here"}

# List completed executions
response = httpx.get(
    f"{API_URL}/models/executions/all",
    params={"status": "completed", "page_size": 5},
    headers=headers,
)
data = response.json()
for execution in data["items"]:
    print(f"Status: {execution['status']}")
    print(f"Objective: {execution['result_data']['objective_value']}")

Response

{
  "items": [
    {
      "id": "a3f8b2c1-1234-5678-abcd-ef0123456789",
      "model_id": "b9c1d2e3-abcd-1234-5678-f0a1b2c3d4e5",
      "status": "completed",
      "result_data": {
        "status": "optimal",
        "objective_value": 1250.50,
        "variables": [
          {"name": "east_allocation", "value": 320},
          {"name": "west_allocation", "value": 180}
        ],
        "solve_time_seconds": 2.34
      },
      "created_at": "2026-02-19T10:05:00Z",
      "completed_at": "2026-02-19T10:05:12Z"
    }
  ],
  "total": 24,
  "page": 1,
  "page_size": 5
}

GET /api/v2/models/executions/{execution_id}

Get the status and result of a specific execution.

Authentication: Requires API key or JWT token.

Path Parameters

Response

Examples

import httpx
import time

API_URL = "https://jaot.io/api/v2"
headers = {
    "Authorization": "Bearer ok_live_your_key_here",
    "Content-Type": "application/json"
}

# Execute a model
response = httpx.post(
    f"{API_URL}/models/a3f8b2c1-1234-5678-abcd-ef0123456789/execute",
    headers=headers,
    json={"input_data": {"warehouses": warehouses, "products": products}},
)
execution = response.json()

# Poll for completion
while execution["status"] in ("pending", "running"):
    time.sleep(2)
    response = httpx.get(
        f"{API_URL}/models/executions/{execution['id']}",
        headers=headers,
    )
    execution = response.json()
    print(f"Status: {execution['status']}")

print(f"Result: {execution['result_data']['objective_value']}")

GET /api/v2/models/executions/{execution_id}/exact-analysis

Exact, solution-based analysis of a completed execution — binding constraints, per-constraint slack and utilization, and objective-term contributions, all computed on demand from the stored solution x* and problem data. Unlike LP-relaxation shadow prices, these figures are exact for the integer solution and identical across solvers. See Analyzing Results.

Authentication: Requires API key or JWT token.

Response

POST /api/v2/models/{model_id}/execute

Execute one of your model projects with input data — model_id is the project's ID (e.g. a marketplace model you forked with Use in studio). All executions run async under the hood; the endpoint waits briefly and returns the result, or a task envelope to poll. For solving a project's draft or a committed version directly, prefer POST /api/v2/projects/{project_id}/solve (see Models); it also supports solution_filter=nonzero for a compact solution.

Real-Time Status Updates

For real-time execution progress, connect to the WebSocket endpoint:

wss://jaot.io/api/v2/ws/executions/{execution_id}

The WebSocket sends JSON messages as the solver progresses:

{"type": "progress", "progress": 0.45, "objective_value": 1234.56, "gap": 0.02}

{"type": "completed", "result": {"status": "optimal", "objective_value": 1250.50, "..."}}

Info: For details on the WebSocket protocol, message types, and connection handling, see the WebSocket documentation.

Errors