P402
P402.io·Meter
DocsPricingDashboard
> _ P402 METER

Meter every AI call before it
becomes budget leakage.

P402 turns each AI request into an economic event with owner, workflow, model, provider, tokens, cost, budget, policy result, outcome, and evidence status.

Install MeterSee dashboard

No prompt storage required. Metadata-only mode is available from the first event.

The problem

AI spend arrives too late.

Provider invoices show totals after usage happens. They do not show which customer, feature, workflow, department, employee, model, or retry caused the cost. P402 records that ownership when the event happens.

What Meter records
Owner
Customer, feature, workflow, department, employee, or API key.
Workflow
The task or process the call served.
Model
Requested and used. Provider and model.
Tokens
Input, output, and total token counts.
Cost
Direct cost in USD, plus retry and context waste.
Budget
Budget the event drew from and remaining balance.
Policy result
Approved, warned, denied, or requires review.
Outcome
Accepted, rejected, revised, escalated, or failed.
Evidence
Receipt and evidence bundle status for audit and finance.
How it works
  1. 01
    Create a P402 key.
    Sign in, generate a key. The key is returned once. P402 stores only a hash.
  2. 02
    Send a request through the OpenAI-compatible endpoint, or report a meter-only event.
    Use the existing OpenAI SDK pattern, or post a metadata-only event from your backend if you call the provider directly.
  3. 03
    Attach customer, feature, department, employee, workflow, or task metadata.
    P402 indexes the event under each owner you supply. Owners are free-form within the documented vocabulary.
  4. 04
    See the event in the ledger.
    The event appears in /dashboard/meter/events with owner, workflow, model, tokens, cost, budget, policy result, outcome, and evidence status.
Privacy

Meter economics, not content.

P402 records economic metadata by default. Prompt and response storage are off by default. Customers choose the privacy mode. Meter-only mode lets your backend call the model provider directly and report the economic event.

  • metadata_onlyOwner, cost, tokens, budget, policy, outcome, and evidence status. No prompts. No responses. Default.
  • fingerprint_onlyAdds a hash fingerprint of prompt and response for dedup and replay protection. Content stays out.
  • redacted_tracePrompt and response retained after a redaction pass for fields the tenant policy allows.
  • private_gatewayYour environment hosts the inference, P402 records the economic event over a signed channel.
  • full_traceOpt-in. Prompts and responses retained verbatim with the event. Requires explicit tenant policy.
Read the trust posture
Proof

Same metering layer, four shipped workflows.

Each vertical demo is a working surface on the same metering layer. Use them to see the event detail, ownership attribution, policy result, and evidence status against a concrete workflow.

Healthcare
Medicaid prior authorization with HIPAA-aligned demo mode and human review gate.
Legal
M&A contract due diligence across an 8-document synthetic data room.
Real estate
Tenant application screening with fraud scoring and HUD fair-housing audit trail.
Enterprise
Department, employee, model, and workflow attribution across one organization.
For developers

Know the cost and margin of every AI feature.

Wrap your AI calls once. P402 attributes every token to a customer, feature, workflow, and budget, with cost, outcome, and evidence on every event. Track AI COGS per feature without rebuilding your billing pipeline.

Start free
For enterprise

Give finance a real AI usage ledger.

See AI spend by department, employee, workflow, model, vendor, customer, outcome, and evidence status. Export evidence for finance, procurement, and audit review without exposing prompts.

Run AI Spend Audit
FAQ
Does P402 need prompt content?+

No. The default privacy mode is metadata-only. P402 records owner, workflow, model, provider, tokens, cost, budget, policy result, outcome, and evidence status, without storing prompts or responses.

Does Meter require settlement?+

No. Meter works before settlement. Settlement is added when AI work becomes payable. You can run P402 as a metadata-only ledger and add receipts or settlement later.

Is P402 a replacement for OpenRouter?+

No. OpenRouter is an execution route for LLM calls. P402 wraps any provider, including OpenRouter, and adds attribution, budgets, policy, outcomes, and evidence around each event.

How is this different from LLM observability?+

LLM observability records traces of prompts, responses, latency, and errors. P402 records the economic event around each call: who owned it, which budget it drew from, which policy result it produced, and what evidence exists for it.

How is this different from FinOps?+

FinOps reports cloud spend after the fact. P402 records cost and attribution at the moment of the AI call, so finance, engineering, and compliance see the same event in the same ledger.

What is the first integration step?+

Create a P402 key, then either route an existing OpenAI-compatible request through P402 or send a meter-only event from your backend. The first event lands in the ledger and the page renders with real data.

Get started

Give every token an owner.

Start with one metered event. Metadata-only by default. Add receipts and settlement when the work becomes payable.

Install MeterSee dashboard