Frequently Asked Questions

Claudexia is a Claude API gateway that gives you pay-as-you-go access to Anthropic's Claude models. No Anthropic account is required — just sign up, get an API key, and start making requests.

Claude (Opus 5.5, Opus 5, Sonnet 5, Sonnet 5.5, Fable 5, Haiku 4.5 and earlier Opus and Sonnet versions), GPT-6 and GPT-5.6, Gemini, DeepSeek, Qwen, GLM, Grok, Kimi and MiniMax. For images there are GPT Image and Nano Banana, for video Veo, Kling, Seedance and more. The current list with prices is on the Models page.

No. Claudexia provides its own API keys. You do not need a separate Anthropic account or API key.

Retention and training terms depend on the API supplier and recipient chain. Verification is not complete, so no general no-storage or no-training guarantee is made. Review the data and recipient documents in the legal section first.

Both the Anthropic Messages API format and the OpenAI Chat Completions format are supported. Use whichever fits your existing code.

You pay per million tokens, with different rates for each model. See the Pricing page for the full table.

No. Your balance never expires.

The minimum top-up is $1. The maximum for a single payment is $10,000. The top-up dialog shows the exact limits for the payment method you choose.

YooKassa (Russian bank cards, SBP), CryptoCloud (cryptocurrency), CryptoBot (Telegram), and Platega (SBP QR, cards, international cards, crypto).

Yes. Both CryptoCloud and CryptoBot accept cryptocurrency payments.

For SDKs using the OpenAI format, use https://api.claudexia.tech/v1. For Claude Code or the Anthropic SDK, use https://api.claudexia.tech (without /v1).

Yes, but you need a paid Cursor plan. Turn on the OpenAI-compatible mode, set the address https://api.claudexia.tech/v1 and use the Cursor model name, for example cursor5. The name table is in the Cursor guide.

Yes. Set ANTHROPIC_BASE_URL to https://api.claudexia.tech (no /v1), put your key in ANTHROPIC_AUTH_TOKEN and the models in ANTHROPIC_DEFAULT_*. A ready settings.json is in the Claude Code guide.

Yes. Point the OpenAI SDK's base_url to https://api.claudexia.tech/v1 and use your Claudexia API key.

Claudexia API keys start with sk_cdx_.

Claude Sonnet 5 is enough for everyday code: it is fast and inexpensive. If you need the highest quality, use Claude Opus 5.5. Our Codex guide uses GPT-6 Sol by default.

Yes. All Claude models support image input.

Extended thinking is available on Opus and Sonnet models. Enable it by passing a thinking parameter in your request.

Most text models have 1M tokens: Claude Opus 5.5, Sonnet 5, GPT-6 and others. Claude Haiku 4.5 has a 200K context and Grok models have 256K. The exact value for each model is in the catalog.

Premium gives all models. Cheap costs less and charges one price for any token, but contains only some models. The channel is set per key in the API Keys section. Details are on the Premium and Cheap channels page.

Yes. POST /v1/images/generations creates images and POST /v1/videos creates video. Money is charged only for a successful result. Details are on the Images and video page.

Provider account or channel policy may limit requests. The dashboard key API does not expose per-key RPM, TPM, or concurrency settings and publishes no numeric quota defaults. Handle HTTP 429 in your client.

Register at claudexia.tech, then go to Dashboard → API Keys to create one.

Yes. You can create as many API keys as you need.

A key supports name, Premium or Cheap channel, optional Premium fallback for Cheap, expiry, and usage statistics. The dashboard key API does not expose per-key RPM/TPM, concurrency, budgets, model allowlists, IP filters, or IDE restrictions.

Yes. TOTP-based two-factor authentication is available in your account settings.

Create a new key in the Dashboard, update your applications to use it, then delete the old key.