Pricing
Wholesale is the price. Nothing else to buy.
One balance, every model. Builders pay wholesale. Resellers pay wholesale and charge whatever they like.
- One balance for every model
- Credits never expire
- Agents can pay per prompt in USDC
- Our price next to OpenRouter's, where it lists the model
- Keys, analytics, usage export
- Your domain with automatic HTTPS
- Default spread, per-model overrides, aliases
- Customers and keys, billed to your balance
- Profit export per customer and model
- Free models to try
- No seats, no plans, no monthly fee
- Usage charged at wholesale from your balance
- Cancel by doing nothing
- Volume pricing on request
Prices
The same models. For less.
Our price on popular models next to OpenRouter's list price, as of 2026-10-16 19:52 UTC. Usually far below.
| Model | Context | Our price / 1M in · out | OpenRouter / 1M in · out | vs list |
|---|---|---|---|---|
Claude Sonnet 5 claude-sonnet-5 | 1M | $0.088 · $0.440 | $2.00 · $10.00 | −96% |
Claude Opus 5.5 claude-opus-5.5 | 1M | $1.20 · $5.90 | $4.00 · $20.00 | −70% |
GPT-5.5 gpt-5.5 | 1.1M | $1.30 · $7.50 | $5.00 · $30.00 | −75% |
Gemini 3.5 Flash gemini-3.5-flash | 1.0M | $0.023 · $0.140 | $1.50 · $9.00 | −98% |
DeepSeek V4 Flash 0423 deepseek-v4-flash | 1.0M | $0.064 · $0.190 | $0.091 · $0.182 | above list |
Kimi K3 kimi-k3 | 1.0M | $0.061 · $1.20 | $0.304 · $11.94 | −90% |
gpt-oss-120b gpt-oss-120b | 131K | $0.0011 · $0.0045 | $0.150 · $0.600 | −99% |
GLM 5.3 Flash glm-5.3-flash | 1.0M | $0.0085 · $0.150 | $0.017 · $0.781 | −81% |
Llama 3.3 70B Instruct llama-3.3-70b | 131K | $0.023 · $0.049 | $0.100 · $0.320 | −84% |
Grok 4.7 grok-4.7 | 500K | $1.70 · $5.10 | $2.00 · $6.00 | −15% |
FAQ
Questions, answered.
How are your prices set?
Every request is routed to the cheapest healthy source for that model, across several providers. Popular models usually cost less than OpenRouter's list price; the ones that cost more are marked "above list" on the Models page. Prices can change at any time, and the Models page shows OpenRouter's list price next to ours, so you can compare.
Is it really the same model?
Yes. A model id always means that model and version, never a renamed substitute. Like any API, answers can vary from one run to the next, even at temperature 0.
What do my customers see?
Your brand, your domain, your prices and your model names. Klinn appears nowhere: not in headers, not in error messages, not in the model list, not in an email.
How do I get paid as a reseller?
Directly. Connect a wallet you control and share it with your clients: they pay you in USDC on Base, or however you bill them. Klinn never holds or forwards client money. When a client uses a model, the charge is debited from your balance.
How reliable is it?
Fail-over to the next provider before the first byte when a model has more than one, circuit breakers per provider and model, and a catalog refreshed every five minutes. The latest stress test served 430 of 430 requests at up to 50 concurrent.
Who sees my prompts?
The provider that serves the request processes it under its own terms, and so does any provider it passes the request on to; we store metadata only (model, tokens, latency, cost), never prompt or completion bodies.
Any fees?
No monthly fee and no plan: pay as you go from one balance, by card or in USDC. Credits never expire and cannot be withdrawn. Reselling is free: you pay wholesale, nothing else.
Can AI agents pay by themselves?
Yes. Call the API with no key and it answers 402 with the price of that one completion; the agent pays it in USDC from its wallet (x402) and gets its answer in the same exchange. Agents with an API key can also top up their balance from code when it runs low.
How does an agent find the catalog and the price?
The catalog needs no key: GET https://app.klinn.dev/api/v1/models lists every model with the wholesale price per million prompt and completion tokens. The USDC for one call depends on the prompt and the output cap, so send that call to POST https://api.klinn.dev/v1/chat/completions with no key. The 402 carries a PAYMENT-REQUIRED header (x402 v2) with the exact amount, and a paid response repeats it in x-price-usd.
Do paid requests use the same route and price as a balance?
Yes. A paid request is ranked like a prepaid one: the cheapest healthy provider for that prompt and output cap, then the next provider if the first fails before the first byte. The quote is our price for that prompt and output cap, at the catalog's rates, and at least $0.001. You pay that quote, which is the price of the output cap. If a later provider has to serve the call, the quote stays the same. A request no provider accepts is not settled.
Will a standard x402 client and a Base wallet work?
Yes. The challenge is x402 v2, scheme exact, USDC on Base (eip155:8453), signed as an EIP-3009 authorization. @x402/fetch with ExactEvmScheme reads PAYMENT-REQUIRED, signs with the wallet, and retries with PAYMENT-SIGNATURE. The older X-PAYMENT header is accepted too. The private key stays on the agent. The completion comes back as soon as a provider accepts, with x-price-usd; the transfer is submitted alongside it. Top-ups wait for settlement and return PAYMENT-RESPONSE.
Do paid x402 requests stream?
Yes. "stream": true is server-sent events, the same shape as a request with an API key, including a usage chunk when the provider sends one. The USDC amount is fixed before the first byte from the prompt and the output cap (max_tokens, 512 if you omit it, 4,096 at most), and the provider is held to that cap. The stream starts after the signature verifies and a provider accepts. The usage chunk reports the token count; the charge stays the quote.
Can an agent run a reseller business?
Yes. It signs in with its own wallet and can do everything the console does: set prices, add customers and their keys, connect a domain, read earnings. You can also add your agent's wallet to your own organization and let it run that.
Can I manage Klinn from Claude, Codex or Cursor?
Yes. On Home, create one key and paste it in. Then ask in plain words: call a model, add a customer and give them a key, or show this week's earnings. What the agent changes shows up on Home.
Can I pay with crypto?
Yes: USDC on Base, from any wallet or straight from an agent. Cards work too, through Stripe.