One API, 200+ models
OpenAI-compatible and Anthropic-compatible gateways with transparent passthrough pricing. The aggregator with the workspace built in.
Spend credits at the prices shown in the app. The only fee is a flat 9.5%, charged once when you add credits. Nothing per operation.
Drop-in for clients that already speak OpenAI or Anthropic. Zero refactor to point at idapt.
One quota across every provider. Per-model envelopes still enforced, but you don't track them yourself.
Cap a key by model, per-period spend, or rotation date. Revoke without restarting clients.
How API pricing works
Add credits to your balance, then spend them across idapt's API and app. You spend at the prices shown in the app. The only fee is a flat 9.5% platform fee when you add credits; there are no other fees and nothing per operation.
Need more than your plan includes?
Subscribers can add Usage Credits and pay only for what they use beyond the included allowance, at the prices shown in the app. Credits never expire.
Don't want a plan? Bring your own API key and use your own provider account.
OpenAI compatible
https://idapt.app/v1
Point any OpenAI SDK at this base URL with your idapt API key. Models from every provider are reachable via OpenAI-shape request/response objects.
Anthropic compatible
https://idapt.app/anthropic/v1
Drop-in for the Anthropic SDK and Claude-Code. Cross-provider models translated at the wire level — your client keeps speaking Anthropic.
FAQ
What does the API cost?
You add credits, then spend them across idapt. Everything is billed at the price you see in the app. The only fee is a flat 9.5% platform fee charged once, when you add credits. There are no other fees.
Is the API OpenAI-compatible?
Yes. Point the OpenAI SDK at `https://idapt.app/v1` with your idapt API key and every OpenAI-shaped client (LangChain, LiteLLM, the official `openai` packages, etc.) works unchanged. Anthropic-shape clients can talk to `/anthropic/v1` instead.
Do I need separate billing for every provider?
No. One credit balance, every provider. We handle the per-provider rate limits and quota allocation behind the scenes — your API key gets a unified rate-limit envelope.
How do I get an API key?
Sign in, open Settings → API Keys, and mint one. Keys can be scoped to specific models, capped per-period, and rotated without restarting any client.
What's the difference vs OpenRouter?
Same idea (one API, many providers) but idapt adds: a full workspace (files, agents, computers), and no per-token aggregator fee — a single flat 9.5% fee applies only when you add credits. Plus direct-provider connectors for top models (lower latency than OpenRouter on flagships).
Are there rate limits?
Rate limits scale with your usage and provider quota. Free hits a daily cap; funded usage uses a sliding window keyed on input tokens, output tokens, and request count. See the rate-limiting docs in the API reference for current numbers.
Building a consumer product instead? See consumer plans · Developer docs