API reference

Maximo AI

Business-aware AI grounded in the workspace behind your key — the same Maximo AI that powers MyTabulon.

https://api.mytabulon.com/v1

GET/ai/modelsai.read

List the AI models available on your plan, including standard and long-context token pricing.

response

{
  "object": "list",
  "data": [{
    "id": "gpt-5.6-terra",
    "pricing": {
      "currency": "USD",
      "unit": "per_1m_tokens",
      "input": 2,
      "output": 12,
      "long_context": {
        "input_token_threshold": 272000,
        "threshold_operator": "greater_than",
        "input": 4,
        "output": 18,
        "applies_to": "full_request"
      }
    }
  }]
}

POST/ai/responsesai.responses

Create an AI response (streams by default). Set stream: false for a single JSON payload. Pass session_id to continue a conversation. Limited to 20 requests/minute.

request

{
  "input": "Draft a polite payment reminder for invoice INV-2026-0042",
  "instructions": "Keep it under 80 words",
  "model": "optional-model-id",
  "session_id": "optional-session-to-continue",
  "stream": true
}

response

// If stream is true (default):
// Returns a sequence of OpenAI-compatible data-only Server-Sent Events (SSE):
data: {"id":"airesp_xxx","object":"chat.completion.chunk","choices":[{"delta":{"content":"Hi"}}]}
data: {"id":"airesp_xxx","object":"chat.completion.chunk","choices":[{"delta":{"content":" there!"}}]}
data: [DONE]

// If stream is false:
{
  "object": "ai.response",
  "id": "airesp_xxx",
  "session_id": "session_xxx",
  "model": "maximo-pandora-3.8-nano",
  "output_text": "Hi — just a friendly reminder that invoice INV-2026-0042 ...",
  "duration_ms": 2150
}

GET/ai/usageai.read

Token usage, cost, and long-context request counts by model and by day. Filter: days (1–90).

response

{
  "object": "ai.usage",
  "days": 30,
  "totals": {
    "requests": 84,
    "input_tokens": 932410,
    "output_tokens": 48220,
    "cost_usd": 3.1875,
    "long_context_requests": 2
  },
  "pricing_note": "The provider's full input context selects the pricing tier; that tier applies to the full request.",
  "by_model": [{
    "model": "gpt-5.6-terra",
    "requests": 16,
    "cost_usd": 2.675,
    "long_context_requests": 2,
    "pricing": { "input": 2, "output": 12, "long_context": { "input_token_threshold": 272000, "input": 4, "output": 18 } }
  }]
}