Claude Fable 5.1 API proxy pricing on Vana API

§ One address, three protocols

One address, three protocols, minimal setup. That is the whole idea behind Vana API. Every request goes to https://api.vanaapi.com, and the protocol you speak depends on the model family you want to use. The same endpoint hosts the Anthropic Messages path, the Gemini native path and the OpenAI format path, so there is nothing to stitch together and no separate configuration to maintain per provider.

That address only answers API requests. The dashboard, where keys, balance and usage log live, is on the main site, https://vanaapi.com. Keeping the two apart is deliberate: the request surface stays simple, while the account surface stays in one place where you can create a key, add funds and read what each call cost.

For Claude models, the Anthropic Messages path (/v1/messages) is used. For Gemini models, the Gemini native path (/v1beta/models/...) is used. For the GPT, Grok, DeepSeek, GLM, Kimi, Qwen, MiniMax and MiMo families, the OpenAI format (/v1/chat/completions) is used, and GPT models also answer on /v1/responses. So when you plan a project, you pick a model and you follow the matching protocol. Vana API keeps the address constant while the route changes with the model.

This is the part that saves time in practice. There is no gateway to configure per family, no base URL to rewrite when you switch models, and no client list to reconcile. One address, three protocols, and a short decision to make about which model you actually want to call.

§ Setting up Claude Fable 5.1 in a few minutes

Claude Fable 5.1 on Vana API is used only from official Claude Code or Claude Desktop, version 2.1.285 or later. Those applications speak the Anthropic Messages protocol themselves, so they connect to https://api.vanaapi.com directly and send requests along /v1/messages. No other route is needed.

The setup sequence is short. Sign up on the main site, create a key in the dashboard, add funds to your prepaid balance, and point your Claude Code or Claude Desktop installation at the Vana API address using the key you just created. That is the whole configuration. There is no account pool to activate, no regional workaround to apply and no separate subscription to sign up for with a model vendor.

Once the address is in place, the model id is claude-fable-5-1. That is the identifier you select in the client when you want Claude Fable 5.1, and it is the identifier Vana API bills against. If you are switching between Claude models later, the change is a model id change rather than a new integration.

Because the client handles the protocol, the setup is genuinely small. You are not writing a request layer by hand, and you are not translating message formats. The application already knows how to talk Anthropic Messages; Vana API makes that request land on a single, well-known address with a prepaid balance behind it.

§ How billing works: prepaid balance, real tokens

Billing on Vana API is pay-as-you-go on a prepaid balance. There is no monthly fee for the pay-as-you-go route, and there is no subscription tied to a seat or a plan you must renew. You add money, you spend it per call, and the balance shows what is left.

Charges are made per real token, input and output, at the published rate for the model you used. That means the number on the invoice follows the actual work your request generated. A short prompt with a short reply costs less than a long context with a long completion, and you can check both in the dashboard.

Every call is listed in the usage log in the dashboard. The log is where you confirm that a request was accepted, which model served it, and what it cost. If a number on your balance does not match your expectation, the usage log is the place to look, call by call, rather than guessing from a monthly summary.

Two smaller details are worth knowing early. New accounts receive $0.10 of credit as a sign-up gift, which is enough to make a first call and see the flow end to end. And if you bring people in, referral earnings are 14% of what people you invite pay, 5% one level further, with withdrawals available from $13.89. The referral structure is separate from your own token charges, so inviting colleagues does not change how your own calls are billed.

§ Claude Fable 5.1 pricing on Vana API

Claude Fable 5.1 is priced at $3.33 / $16.74 per million tokens for input and output on Vana API. The official list price for the same model is $10 / $50. That gap is the practical reason to look at the Vana API rate sheet before you send volume through a relay service.

For context, the other Claude models on Vana API follow the same pattern. Claude Haiku 4.5 is $0.3472 / $1.60 against an official list price of $1 / $5. Claude Opus 5.5 is $1.46 / $6.94 against $4 / $20. Claude Sonnet 5.5 is $1.18 / $5.76 against $2 / $10. So you can match the model to the task and still see a lower per-token figure on Vana API than the published list price.

Because billing is per real token, these figures translate directly into your own usage. As an example, say you run a workflow every day that sends 2 million input tokens and produces 500,000 output tokens through Claude Fable 5.1. At the Vana API rate, the input side is $6.66 and the output side is $8.37, for a total of $15.03 for that day. This is an example, not a quote, and your own log will reflect whatever your traffic really looks like.

The prepaid balance also means you can start small. The $0.10 sign-up credit lets you confirm the connection works, the usage log lets you measure the first real call, and only then do you decide how much to top up. There is no commitment to make before you have seen the numbers.

§ Monthly plans for Claude, and refunds

If you prefer a fixed monthly arrangement, Vana API offers Claude-only plans that are billed for a 30-day period. The Claude up to $450 / 30 days plan costs $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 costs $153.61 for 30 days, with at most $320 a week, also Claude models only.

Both plans keep the same simple shape: one price for the period, a stated ceiling per week, and a stated ceiling across the 30 days. That makes it easier to decide which one fits your traffic than a flexible but hard-to-forecast arrangement would be. If your usage is irregular or you are still evaluating, the prepaid per-token route remains available on the same address.

Refunds are straightforward as well. Unused balance can be refunded within 3 days of the top-up, minus what was used and the payment processor's fee. The window is short and the deduction is explicit, so it is worth checking your balance and the usage log during the first few days after a top-up rather than assuming every added dollar is still refundable later.

§ Keeping the account side tidy

The dashboard at https://vanaapi.com is where the practical work happens. Keys are created there, so you can issue a fresh key without waiting on anything else. Balance is shown there, so you can see what remains before a call turns into a surprise. The usage log is there, so every request is accounted for line by line.

Together with one address at https://api.vanaapi.com, that gives you a short, repeatable routine: create a key, add funds, select the model id, read the log. For Claude Fable 5.1 specifically, the model id is claude-fable-5-1, and the client is official Claude Code or Claude Desktop, version 2.1.285 or later, speaking Anthropic Messages to the same address as every other family.

One address, three protocols, minimal setup. Vana API is built so that the part you configure once stays small, and the part you check daily, the balance and the usage log, is right there in the dashboard.