Stop AI spend before it's charged.
Prove every call after.

• Cost attribution is broken

  The bill arrives as one number. Which team? Which feature? Which customer? Nobody can answer the CFO's question.

• AI agents have no governance

  Your AI SDR, support bot, and coding agent each generate thousands of requests daily. None of them appear on one ledger.

• There's no proof of what happened

  When auditors ask for AI usage records, you have provider invoices and Slack screenshots. That's not a financial record.

1. 01

   Tamper-Evident Ledger

   Every AI request TOLVYN meters is hash-chained and HMAC-signed. Cryptographic proof, not logs. Your CFO or auditor can re-verify the hash chain themselves with the script shipped in every evidence package — the HMAC is keyed on a secret TOLVYN holds, so it detects tampering by anyone without that key.

2. 02

   Per-End-Customer Attribution

   Track AI cost per your customer. Know your AI COGS per customer. The unit economics SaaS founders are missing.

3. 03

   Invoice Reconciliation

   Upload your OpenAI / Anthropic / Google usage export. TOLVYN shows the gap: spend on your provider bill that did not go through TOLVYN.

4. 04

   Agent Governance

   Per-agent budgets. Cost anomaly alerts — one call over 10× the recent average. Kill switch. Every AI agent on one ledger.

5. 05

   Savings Analyzer

   Nightly analysis identifies migration opportunities. Switch to a cheaper model where your actual usage patterns fit.

6. 06

   Pricing Change Alerts

   When OpenAI changes prices, customers using affected models get an alert projecting the monthly impact from their last 30 days of usage.

• Budgets per end customer. End customers are attributed and reported, but a budget can only be set on the organization, a team, a service, an agent or an API key.
• Limits per agent run or task. Budgets and quotas are per period; there is no per-run limit yet.
• Amazon Bedrock, Azure OpenAI and Vertex AI. The providers TOLVYN proxies natively are OpenAI, Anthropic, Google's Gemini API and DeepSeek.
• Single sign-on. There is no SAML or OIDC sign-in yet; you sign in with email and password.
• A SOC 2 report. There is none yet. What an auditor gets today is the evidence package and the ledger it verifies.
• Gemini's Files API upload and Gemini's OpenAI-compatible endpoint. Both are answered with a named 501. Upload files to Google directly and pass the file reference through TOLVYN.
• A fallback in proxy mode. If TOLVYN is unreachable, a base-URL integration fails. Fail-open lives in the SDKs and needs your own provider key.
• More than one region. TOLVYN is hosted in India only.

# Before
from openai import OpenAI
client = OpenAI(api_key="sk-...")

# After
from tolvyn import OpenAI
client = OpenAI(
    tolvyn_api_key="tlv_live_...",
    openai_api_key="sk-...",      # REQUIRED for fail-open; no key, no fallback
    team="engineering",
    service="chatbot-api"
)
# Everything else identical

// Before
import OpenAI from 'openai'
const client = new OpenAI({ apiKey: 'sk-...' })

// After
import { OpenAI } from 'tolvyn'
const client = new OpenAI({
  tolvynApiKey: 'tlv_live_...',
  openAIApiKey: 'sk-...',        // REQUIRED for fail-open; no key, no fallback
  team: 'engineering',
  service: 'chatbot-api'
})

# Before
curl https://api.openai.com/v1/chat/completions \
  -H "Authorization: Bearer sk-..."

# After (change the URL, keep everything else)
curl https://proxy.tolvyn.io/v1/proxy/openai/v1/chat/completions \
  -H "Authorization: Bearer tlv_live_..." \
  -H "X-Tolvyn-Team: engineering" \
  -H "X-Tolvyn-Service: chatbot-api"

pip install tolvyn  ·  npm install tolvyn  ·  go get github.com/tolvyn/tolvyn-go

Works with OpenAI, Anthropic, Google, and DeepSeek. If TOLVYN is unreachable, SDK mode automatically retries against the provider directly.