Vana API pricing: GPT-6 Astra at $0.6944 / $3.54

Money mattersEN2026-11-06

§ One address, three protocols, minimal setup

Vana API gives you a single place to send model requests: api.vanaapi.com. That address speaks three protocols, so the endpoint you use depends on the model and the client you already have. There is little to set up. You create a key in the dashboard on vanaapi.com, add credit to your prepaid balance, and point your client at the one address.

The important distinction is which client speaks which protocol. Claude models are available only from official Claude Code or Claude Desktop, version 2.1.285 or later, because those applications speak Anthropic Messages (/v1/messages) themselves. Gemini models are available only from Gemini CLI, over the Gemini native protocol (/v1beta/models/...). GPT, Grok, DeepSeek, GLM, Kimi, Qwen, MiniMax and MiMo models use the OpenAI format (/v1/chat/completions). GPT models also answer on /v1/responses, which is what Codex uses.

This arrangement keeps the setup narrow and predictable. Vana API is not trying to make every model look the same to every application. Instead, you use the protocol each model expects and keep the base address unchanged. The api.vanaapi.com address answers API requests only. Keys, balance and the usage log live in the dashboard on vanaapi.com.

Claude models: official Claude Code or Claude Desktop, version 2.1.285 or later
Gemini models: Gemini CLI, using the Gemini native protocol
GPT, Grok, DeepSeek, GLM, Kimi, Qwen, MiniMax and MiMo: OpenAI format, including /v1/responses for GPT

§ GPT-6 Astra beside its official list price

GPT-6 Astra is listed on Vana API at $0.6944 per million input tokens and $3.54 per million output tokens. Its official list price is $10 per million input tokens and $50 per million output tokens. Put side by side, the comparison is straightforward: Vana API charges $0.6944 where the official list price shows $10 for input, and $3.54 where the official list price shows $50 for output.

The model id is gpt-6-astra. Because GPT models use the OpenAI format, you can send GPT-6 Astra requests from Cursor, Cline, Roo Code, OpenCode, Cherry Studio, any OpenAI SDK or cURL. GPT models also answer on /v1/responses, which is what Codex uses. The same base address covers these routes, so switching between supported GPT clients does not require a new provider setup.

For example, say you run a workflow that sends 2,000,000 input tokens and 200,000 output tokens through GPT-6 Astra in a month. At Vana API rates, the input cost starts from $0.6944 multiplied by 2,000,000, and the output cost starts from $3.54 multiplied by 200,000. The official list price for the same token counts starts from $10 multiplied by 2,000,000 and $50 multiplied by 200,000. This is an example of arithmetic, not a quote for your own workload.

Several other GPT entries in the price list are lower again. GPT-6.1 Sol is $0.1389 / $0.8333 on Vana API, against an official list price of $2 / $10. GPT-5.6 Sol is $0.3472 / $1.81, against $4 / $20. Vana API therefore gives you a small set of GPT options with visible per-million-token pricing rather than a single undifferentiated rate.

§ Prepaid billing with no monthly fee

Vana API uses a prepaid balance. You are charged per real token, and there is no monthly fee. That makes the cost easier to follow when you are starting out, because you can see the balance in the dashboard and watch individual calls appear in the usage log. Every call is listed there, so the record is attached to the account you already set up.

The order of operations is short. Create your key in the dashboard, top up the balance, and use the key with your chosen client. New accounts receive $0.10 of credit as a gift on sign-up. That is enough to make a small first request and see how the key, address and protocol fit together before committing to a larger top-up.

Refunds are available under a clear condition: unused balance can be refunded within 3 days of the top-up, minus what was used and the payment processor's fee. If you are still learning the setup, that window gives you time to test the connection and check the usage log while the balance remains unused.

For regular use, monthly plans are also available. The Claude up to $450 / 30 days plan is $64.72 for 30 days, with at most $140 a week, and it covers Claude models only. The Claude up to $1,050 / 30 days plan is $153.61 for 30 days, with at most $320 a week, also for Claude models only. The GPT up to $2,750 / 30 days plan is $103.61 for 30 days, with at most $820 a week, and it covers GPT models only.

Each plan states its own scope, so check that it matches the models you intend to call. A prepaid balance works without a monthly commitment, while the plans are there when you prefer a defined 30-day arrangement.

§ Using the OpenAI format for GPT models

For newcomers, the most practical route is usually the one their existing tools already support. GPT models work over the OpenAI format at /v1/chat/completions, and they also answer on /v1/responses. That means you can work from Cursor, Cline, Roo Code, OpenCode, Cherry Studio, any OpenAI SDK or cURL, with the base address set to api.vanaapi.com and the model set to gpt-6-astra.

There is no need to convert your workflow into a different protocol just to try the pricing. The model id is what you select; the endpoint determines the format. When you compare cost, use the per-million-token figures for the exact model you choose. GPT-6 Astra at $0.6944 / $3.54 and GPT-6.1 Sol at $0.1389 / $0.8333 are separate entries with separate rates.

The usage log in the dashboard gives you a place to confirm what was actually spent. Since billing follows real tokens, a short request and a long request will not be treated as equal. Watching the log after your first few calls is a simple way to understand which part of your workflow drives the bill.

Referral earnings follow a similar visible structure. You receive 14% of what people you invite pay, and 5% one level further. Earnings can be withdrawn from $13.89. Referral credit is separate from the prepaid balance used for model calls, so the two should not be confused when you review your account.

§ Keeping the dashboard and the API address straight

A small point that saves time: api.vanaapi.com is the address for API requests. The dashboard is on vanaapi.com, where you create keys, view your balance and read the usage log. Keeping those two roles separate makes the setup easier to explain and easier to maintain.

Vana API's approach is deliberately compact. One address, three protocols and minimal setup are the core of the service. You choose a supported model, use the client that speaks its protocol, and check the dashboard when you want to see balance and calls. For GPT-6 Astra, the price comparison with the official list price is $0.6944 / $3.54 on Vana API against $10 / $50 officially.

That is the whole starting point: a key, a prepaid balance, one address and the right protocol for the model in front of you.