One API for every AI model
TokenRoute routes your requests to 458+ models through a single OpenAI-compatible endpoint, billed per request at 30% below provider pricing, with per-key spend limits and full usage analytics.
No subscription · pay only for what you use · works with the OpenAI SDK
- models
- 458+
- models
- providers
- 8+
- providers
- compatible
- OpenAI
- compatible
- below list price
- 30%
- below list price
curl https://tokenroute.app/api/v1/chat/completions \
-H "Authorization: Bearer $TOKENROUTE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-4o-mini",
"messages": [{ "role": "user", "content": "Hello!" }]
}'Access frontier models from the labs you already trust
OpenAI
Anthropic
Google
Meta
Mistral
DeepSeek
Qwen
xAIOpenAIAnthropicGoogleMetaMistralDeepSeekQwenxAIFeatures
Everything you need to ship with AI
A complete gateway: integrate once, switch models freely, and only pay for what you use.
Every model, one endpoint
Hundreds of models from every major provider behind a single, unified API.
OpenAI-compatible
Drop-in with the OpenAI SDK. Just change the base URL and your API key.
30% below list price
Every model at 70% of the provider's own per-token rate. No subscription, no lock-in.
Usage analytics
Per-request token and cost tracking, broken down by model and API key.
Built-in playground
Test any model in an interactive chat with live token and cost readouts.
Keys & limits
Create scoped API keys with per-key spend limits and instant revocation.
Powered by the best
458+ models, ready to call
Frontier chat, reasoning, vision, and open-source models from the labs you already trust, all reachable with the same request. Switch models by changing one string.
OpenAIAnthropicGoogleMetaMistralDeepSeekQwenxAIProvider prices, minus the markup
Same models, same quality - at 70% of provider list price. Output cost per 1M tokens, side by side.
Simple, usage-based pricing
Provider-grade models, 30% below provider pricing. You only pay for the tokens you use.
Pay-as-you-go
30% below provider pricing
No subscription · start at any balance
- Pay-as-you-go - billed per request, to the token
- No monthly subscription or seat fees
- Start at any balance, with no minimum spend
- Per-key spend limits and real-time usage tracking
FAQ
Frequently asked questions
Everything you need to know about the API, models, pricing, and payments.
What is TokenRoute?
TokenRoute is an AI API gateway that routes your requests to hundreds of large language models from providers like OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, Qwen and xAI through a single OpenAI-compatible endpoint. Every model is priced below that provider's own published per-token rate, billed per token with no subscription and no minimum spend.
How is TokenRoute different from calling AI providers directly?
Instead of integrating each provider's SDK, key and billing separately, TokenRoute gives you one API key and one OpenAI-compatible endpoint for every model. You switch models by changing one string, get unified per-request usage and cost analytics, and pay from a single balance.
Is TokenRoute compatible with the OpenAI SDK?
Yes. TokenRoute exposes an OpenAI-compatible /v1/chat/completions API. Point your existing OpenAI client's base URL at TokenRoute and use your TokenRoute API key. No other code changes are required. Streaming via Server-Sent Events is supported.
How does TokenRoute pricing work?
Pricing is pay-as-you-go and billed per token, to the request. There is no monthly subscription or seat fee. You only pay for what you use, at or below the underlying provider's pricing.
How do I pay for and top up my TokenRoute balance?
You add credits from the Credits page, in any amount. Checkout is handled by a payment provider, and your balance is credited automatically once the payment confirms. There is no subscription: credits are spent per token as you make requests.
Which AI models and providers does TokenRoute support?
TokenRoute routes to hundreds of models across chat, reasoning, vision and open-source families from providers including OpenAI, Anthropic, Google, Meta, Mistral AI, DeepSeek, Qwen, xAI, and more. Browse the full, live catalog with per-token pricing at /models, with no account required.
Do I need a subscription to use TokenRoute?
No. TokenRoute has no subscription and no seat fees. Create an account, add credits, and pay only for the tokens you actually use.
Can I set spending limits and manage API keys?
Yes. You can create multiple scoped API keys, set a per-key spend limit, track usage per key and per model in real time, and revoke any key instantly.
How do I get started with TokenRoute?
Create a free account, generate an API key, add credits, then set your OpenAI client's base URL to TokenRoute and make a request. Most developers are live in a few minutes.
Ship your AI features today
Create an account, generate a key, and make your first request to TokenRoute in minutes.