Product
Control, cap and account for every AI API call
A1 API Gateway gives each application, teammate and customer its own key to Anthropic, OpenAI and xAI models, and gives you the limits, the numbers and the audit trail behind every one of them.
Virtual keys
A key for every consumer, with its own rules
Keys are shown once when they are created and stored only as a hash. Each one is named, has an owner, and carries its own limits.
- IP rules per key: an explicit allowlist, allow-anywhere, or pinned to the first address that uses the key.
- Requests per minute, tokens per minute and a cap on concurrent requests.
- Dollar budgets over the key's lifetime or per month, week, day or hour, with the time each one resets.
- Restrict a key to particular models, or leave it to use whatever the organisation has switched on.
- A per-key log of the addresses it was used from and of its requests, with prompt and completion logging available per key when you want it.
Vendor keys and routing
Your vendor accounts, kept out of reach
Organisations can bring their own keys for Anthropic, OpenAI and xAI, several per vendor, told apart by a label such as production or testing.
- Every vendor key is tested against the vendor before it is saved.
- Secrets are encrypted with AES-256-GCM and are never returned by the API or written to logs.
- Pick a default key per vendor, or bind an organisation or a single virtual key to a specific vendor key.
- Spend is attributed to the vendor key that served it, so each vendor account can be reconciled on its own.
Spend and credit
Numbers you can take to finance
Every request is metered, priced in US dollars from the model's token rates, and recorded against the key, the model and the vendor key.
- Own-key requests are billed to you by the vendor; the gateway meters and caps them without deducting anything.
- Prepaid requests run on our vendor keys and are deducted from credit that can never go below zero.
- A monthly ceiling for the whole organisation, on top of each key's budgets.
- Daily reconciliation against Anthropic's and OpenAI's own usage reports, with differences flagged.
Consoles
A console for each organisation
Each organisation signs in at its own address, such as acme.a1apigateway.com. Two-factor authentication is required for everyone.
- A dashboard of spend and requests by period, provider, model, vendor key and virtual key.
- Keys, vendor keys, model selection, aliases, members and usage in one place.
- Owner, admin and viewer roles, and invitations by email.
- Alerts for budgets, the ceiling, credit and vendor-key health, by email or signed webhook.
- An audit log recording who changed what, with the values before and after.
API
Change the base URL and the key
The OpenAI SDK, the Anthropic SDK and plain HTTP all work unchanged. Model names carry the vendor, such as anthropic/claude-sonnet-5, openai/gpt-5 or xai/grok-4.
| Endpoint | What it does |
|---|---|
POST https://api.a1apigateway.com/v1/chat/completions | Chat with any supported model in the OpenAI format, streaming included |
POST https://api.a1apigateway.com/v1/messages | Claude models through Anthropic's native Messages API |
POST https://api.a1apigateway.com/v1/messages/count_tokens | Count input tokens before sending a Claude request |
GET https://api.a1apigateway.com/v1/models | The models this key may use, with what each one accepts |
GET https://api.a1apigateway.com/v1/key | This key's limits, budgets and routing |
Every response carries x-gw-request-id and the model that actually served it in x-gw-model; a successful one also carries its exact cost in x-gw-cost-usd and the budget left in x-gw-budget-remaining-usd. Errors come back in the envelope your SDK already understands.
See it with your own workloads
We can walk you through it, or you can get started straight away.