API v1hldgroup.org

Usage & audit

Track token consumption, request volumes, and per-model breakdown — plus a filterable audit log of every Axon API call made by your tenant.

Usage summary

bash
GET /v1/axon/usage?days=30
NameTypeRequiredDescription
daysintegerNoLookback window (1–365). Defaults to 30.
json
{
  "data": {
    "period_days": 30,
    "since": "2025-05-02T00:00:00Z",
    "totals": {
      "requests": 4820,
      "prompt_tokens": 9240180,
      "completion_tokens": 1830440,
      "total_tokens": 11070620
    },
    "by_model": {
      "axon-sovereign-1": {
        "requests": 1240,
        "prompt_tokens": 8100200,
        "completion_tokens": 1800000,
        "total_tokens": 9900200
      },
      "axon-embed-1": {
        "requests": 3580,
        "prompt_tokens": 1139980,
        "completion_tokens": 0,
        "total_tokens": 1139980
      }
    },
    "by_operation": {
      "completion": 1240,
      "embedding": 3380,
      "rag_query": 200
    }
  }
}

Audit log

bash
GET /v1/axon/audit
NameTypeRequiredDescription
filter[operation]stringNocompletion | embedding | rag_query
filter[model]stringNoaxon-sovereign-1 | axon-sovereign-1-mini | axon-embed-1
filter[knowledge_base_id]stringNoFilter to a specific knowledge base.
filter[created_after]stringNoISO 8601 datetime lower bound.
filter[created_before]stringNoISO 8601 datetime upper bound.
page / per_pageintegerNoPagination. Max per_page=100.
json
{
  "data": [
    {
      "id": "cmp_01hxyz",
      "tenant_id": "ten_01hxyz",
      "user_id": "usr_01hxyz",
      "model": "axon-sovereign-1",
      "operation": "completion",
      "prompt_tokens": 286,
      "completion_tokens": 201,
      "knowledge_base_id": null,
      "created_at": "2025-06-01T10:00:00Z"
    }
  ],
  "pagination": { "page": 1, "per_page": 20, "total": 4820, "has_more": true }
}
Note:The audit log retains records for 12 months. For longer retention (required by some compliance frameworks), export regularly via the API or contact HLD to configure automated audit log forwarding to your SIEM.

Building a cost allocation report

typescript
// Monthly token cost report by operation type
const usage = await fetch('/v1/axon/usage?days=30').then(r => r.json())
const { by_model, by_operation } = usage.data

// Axon pricing (contact HLD for your contracted rates)
const RATES = {
  'axon-sovereign-1':      { input: 0.015, output: 0.075 },  // per 1K tokens
  'axon-sovereign-1-mini': { input: 0.003, output: 0.015 },
  'axon-embed-1':          { input: 0.0001, output: 0 },
}

const cost = Object.entries(by_model).reduce((total, [model, stats]) => {
  const rate = RATES[model] ?? { input: 0, output: 0 }
  return total
    + (stats.prompt_tokens / 1000) * rate.input
    + (stats.completion_tokens / 1000) * rate.output
}, 0)

console.log(`Estimated Axon cost (30 days): $${cost.toFixed(2)} AUD`)
Data governanceReference: Errors