RelayVano in Hindi: Claude Code and Pay-as-You-Go API Relay

Hindi Pages for a Relay Built for Developers Worldwide

A common assumption about an API relay is that a Hindi-speaking developer will have to work through an English dashboard, guess what the pricing table means, and translate the documentation in their head before writing a single request. RelayVano is built differently. We are an independent relay made for developers worldwide, and the whole site, dashboard and docs come in six languages: 中文, English, Español, हिन्दी, Português, Français.

That includes the pages you actually use. The model pricing tables, the setup notes, the usage history and the dashboard controls are available in हिन्दी, so the experience does not stop at a translated landing page. You can read a fare, add funds, inspect a charge and open a support message without switching to English mid-task.

Prices show in yuan on the Chinese pages and in approximate dollars on the others, which makes the fare list easy to follow on a Hindi page. New accounts start with $0.40 of credit, so you can open the dashboard, read the हिन्दी docs and send a small test request before deciding how much to top up. RelayVano is designed to be picked up quickly, and the language is only the first part of that.

One API Key, Three Protocol Endpoints

Another common assumption is that reaching several model families means juggling several accounts, several keys and several billing dashboards. With RelayVano, one API key reaches the models we serve through https://api.relayvano.com, and the endpoint you use depends on the protocol the model expects.

The OpenAI-compatible path is /v1. Anthropic-style requests use /v1/messages. Gemini models use the native Gemini path at /v1beta, which is the native Gemini protocol rather than an OpenAI-compatible translation. You set your base address and your key in the client you already use, choose the model you want, and RelayVano handles the routing and the charge.

That structure is useful for a team where different people prefer different tools. One member may be working in Claude Code, another in Gemini CLI, and a third may be calling GPT or DeepSeek models from a script. The key stays the same across those paths, and the billing view stays in the same dashboard. For a reader who wants the whole picture in हिन्दी, the docs describe each endpoint in the same six-language format.

Client Pairing Without Guesswork

The most frequent misunderstanding in relay setups is the belief that any client can talk to any model. With RelayVano the pairing is fixed, and knowing it in advance saves an afternoon of failed calls and confusing error messages.

• Claude models work only in the official Claude Code and Claude Desktop clients, version 2.1.285 or newer. They are not paired with Codex, Cursor, Cherry Studio, Cline, OpenCode, OpenAI SDKs or cURL.
• Gemini models use the native Gemini protocol only, in Gemini CLI. They are not OpenAI-compatible, so an OpenAI-style request shape is not the right way to reach them.
• Codex, Cursor, Cherry Studio, Cline, OpenCode, OpenAI SDKs and cURL pair only with GPT, DeepSeek and the other non-Claude, non-Gemini models, including Qwen, Kimi, MiniMax, Hunyuan, Grok and MiMo models.

In practice, this means the model choice and the client choice travel together. If you want Claude Sonnet 5.5, open Claude Code or Claude Desktop at version 2.1.285 or newer and point it at RelayVano. If you want Gemini 3.1 Pro, use Gemini CLI and the native path. If you prefer Cursor or cURL, choose a GPT, DeepSeek, Qwen or similar model instead. RelayVano documents the pairing clearly, and the हिन्दी docs carry the same detail as the English ones.

What Pay-as-You-Go Pricing Costs in Practice

The assumption that relay pricing is vague tends to dissolve once the numbers are visible. RelayVano sells prepaid balance charged per token, with no monthly fee, and the balance never expires. Prompt tokens, reply tokens, cache stores and cache hits are priced separately, and every request shows up with its charge on the usage page.

For a quick look at some fares, Claude Sonnet 5.5 is $1.25 prompt / $6.25 reply / $0.0889 cache hit, compared with an official list price of $2 / $10 / $0.1 cache hit. Claude Haiku 4.5 is $0.3958 prompt / $1.90 reply / $0.0319 cache hit. On the GPT side, GPT-6.1 Sol is $0.1958 prompt / $0.9792 reply / $0.0389 cache hit. Gemini 3.1 Pro is $1.58 prompt / $9.13 reply / $0.1569 cache hit, and MiniMax M3 is $0.1472 prompt / $0.5889 reply / $0.0292 cache hit.

Here is an example only, to show how the arithmetic works. Suppose you send 200 requests a day, each with 1,000 prompt tokens and 500 reply tokens using Claude Sonnet 5.5. The prompt side is 200,000 tokens, which at $1.25 per million tokens is $0.25. The reply side is 100,000 tokens, which at $6.25 per million tokens is $0.625. That example comes to $0.875 per day for that exact mix, before any cache store or cache hit charges. Your own traffic may look very different, which is why the usage page matters.

Because you pay per token rather than on a vendor's subscription, a quiet week does not create a bill you did not plan for, and a busy week scales with what you actually send. There is no monthly fee to cancel, and unused balance stays available.

Top-Ups, Refunds and Referral Credit

Adding money is deliberately uncomplicated. Top-ups go through Stripe checkout, where cards are accepted everywhere, plus WeChat Pay where your account supports it. Unspent top-ups can be refunded within 9 days, so a balance you are not ready to use does not have to feel locked in.

RelayVano also lets you earn from people you invite. You receive 16% of every top-up made by people who sign up through your link, and 4% one level further. Earnings can be withdrawn from $5.56 or moved into your balance, and moving them in is a simple way to fund your own testing without a second payment step.

For a developer who is already explaining RelayVano to Hindi-speaking colleagues, that referral structure turns the recommendation into something useful. The person you help gets a working relay with a हिन्दी interface, and you get a share of their top-ups. New accounts still begin with $0.40 of credit, so the first request does not depend on a referral balance arriving first.

Reliable Charges, Direct Support and Your First Request

One more assumption worth retiring is that an interrupted request is simply lost. If a caller disconnects mid-stream, RelayVano still finishes the request and bills it on the tokens really used. That keeps accounting honest: the work the model completed is the work that appears on the usage page, rather than a request that vanished from both the output and the ledger.

When something is unclear, logged-in members write to us under Messages in the dashboard. You do not need to hunt for a contact form or guess which mailbox reaches the right team. Ask about a fare, a client pairing or a charge, and the conversation stays attached to your account.

RelayVano gives Hindi-speaking developers a relay that reads naturally end to end: the site, dashboard and docs in six languages, one API key across the models we serve, fixed client pairing, per-token pricing you can check request by request, and top-ups that are easy to fund and easy to reverse when they go unused. New accounts start with $0.40 of credit, so the shortest path to an answer is to sign up, read the हिन्दी docs, and send your first request.