Zukko vs the official Anthropic Console

The official Anthropic Console and Zukko run on the same protocol — the Anthropic Messages API. That's where the formal similarity ends; from there the difference is in what you can pay with, which models you can reach, and what happens when something goes wrong.

Payment — the most common blocker

An official Anthropic account is tied to a card that has to clear regional billing checks — in practice that rules out most Russian-issued cards and requires either a foreign card or a middleman service. Zukko accepts SBP (Russia's Faster Payments System), bank cards via PayPear, and cryptocurrency directly, with no third-party workaround needed. Separately, since 2026-09-16, payment via foreign Mastercard EU cards through Fride has been added (a fee of about 13%, amounts $2–$594) — for people who already hold such a card and prefer to pay with it directly.

There's no starting balance on signup at Zukko — that practice was dropped on 2026-09-18 because of disposable-email account farming. A one-time $0.50 bonus is credited for subscribing to the Telegram channel @zukkopro↗, once per Telegram account: link Telegram in the dashboard, subscribe, and press "Check".

One provider, or several under one key

The official Anthropic API only reaches Anthropic's own models. Zukko's single fetch_... key opens several model lines at once: claude-opus-5-5 (Claude Opus 5.5), claude-sonnet-5, gpt-5.6-sol, gpt-5.6-luna, glm-5.3, and glm-5.3-flash. Claude and GPT have a 1,000,000-token context; GLM has 128K and no vision. The current list is at GET https://api.zukko.pro/v1/models.

The practical upshot: if Claude Opus is temporarily unavailable or slower than usual, switching models means changing one model string in the request body — not switching accounts, keys, and SDKs.

Image generation through Zukko is temporarily unavailable — the gateway currently serves text models only.

Request format and tool compatibility

Both platforms support POST /v1/messages with the required anthropic-version: 2023-06-01 header, and both offer a free POST /v1/messages/count_tokens for counting tokens without running the request. For Claude Code, that means switching to Zukko doesn't require swapping the client — only the base URL and the key change.

curl https://api.zukko.pro/v1/messages \
  -H "Authorization: Bearer fetch_your_key" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-sonnet-5","max_tokens":256,"messages":[{"role":"user","content":"hi"}]}'

But Zukko doesn't stop there: the same key also works in OpenAI-compatible mode at https://api.zukko.pro/v1 — POST /v1/chat/completions or POST /v1/responses. If a project uses Claude Code (Anthropic-style) and, say, Cursor or Codex (OpenAI-style) at the same time, there's no need for two separate key sets and two billing accounts — both protocols run through one account.

What's visible in the response, and what can be constrained

Every regular (non-streaming) Zukko response carries X-Zukko-Cost-Micros (the cost of that specific request in microdollars) and X-Fetch-Balance-Cents (the remaining balance after the charge — despite the name, also in microdollars, 1,000,000 = $1) — spend is visible without opening a separate billing dashboard. Streaming responses (stream: true), which Claude Code uses, for example, do not include these headers; there you check spend in the dashboard, /v1/usage, or /v1/balance. An Idempotency-Key header protects against double-charging when the same request gets resent within 24 hours. With an X-Zukko-Cache: true header (temperature: 0 also enables the cache, but only for models that accept it — Claude models may reject a temperature other than 1 with a 400), requests with an exactly matching body get cached for up to 7 days for chat, and a repeat call returns instantly and for free, flagged with X-Zukko-Cache: hit.

Each API key can also be configured individually: a daily and monthly dollar spend limit and a list of allowed models, set right on the key's page in the dashboard. That matters when a key goes to a bot or a contractor rather than yourself — the restriction lives at the key level, not the whole account.

Errors — what the code means

Zukko's error format is predictable and documented by code: 401 for an invalid key ({"error":{"message":"Invalid API key. Use Authorization: Bearer fetch_*","type":"authentication_error"}}), 402 for an empty balance (insufficient_balance, checked before the request reaches the model — a failed request costs nothing), 404 model_not_found for a typo in the model name, 429 when a specific key's limit is exceeded — key_daily_limit_reached (resets at 00:00 UTC) / key_monthly_limit_reached (there is no general rate limit on the gateway; other 429s come from the provider), 403 model_not_allowed_for_key, 422 idempotency_conflict when the same Idempotency-Key is reused with a different request body, 503 provider_unavailable when the provider is down, and 503 model_disabled when a model is switched off.

Support and language

The current model and pricing list is public, no key required, via GET /v1/models and GET /v1/models/info; service and model status is at the public GET https://api.zukko.pro/api/status. The Zukko site is bilingual — Russian on the root domain, English under /en — with the dashboard at zukko.pro/app↗ and keys created at zukko.pro/app/api-keys↗, shown once. Support runs through Telegram at @zukkopro↗, no contact form and no ticket queue. There's also a referral program: 10% of invited users' top-ups.

Bottom line

If all you need is Claude and paying with a foreign card isn't an issue, the official API is the direct path with no intermediary. If paying from Russia without workarounds matters, along with one-key access to several providers and real-time spend visibility straight from response headers, the difference between the console and the gateway is the difference between one access channel and infrastructure built on top of several.