Direct access from Russia — no VPN·USDT / USDC in 5 minutes·Original models, no downgrade

Claude, GPT, Gemini & DeepSeek API in Russia — no VPN, pay in USDT

50+ models with one key. Compatible with the OpenAI and Anthropic APIs — change only base_url. No subscription: pay only for what you use.

No foreign bank card needed USDT / USDC · BEP20 / TRC20 / TON Live in < 5 min
$ curl https://www.ruapi.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5-5",
    "messages": [{"role": "user", "content": "Hello"}]
  }'
Model replyHello! How can I help you today?

Only BASE_URL and KEY — zero business code changes.

// COMPATIBLE WITH
OpenAI OpenAIAnthropic AnthropicGemini Google GeminiDeepSeek DeepSeekKimi Moonshot KimiQwen Alibaba QwenMinimax MiniMaxZhipu Zhipu GLM
# facts

RuAPI in numbers

80
models on one key
10
model developers
$5
minimum top-up, in USDT / USDC or with a foreign-issued card
1–5 min
typical time for USDT / USDC to be credited
7 days
to request a refund of unused balance
$0
monthly fee, subscription or long-context surcharge of our own

New models go live soon after their makers release them; retirements are announced in advance. Dated changelog →

# why teams trust us

Production-grade infrastructure

Three reasons teams across Russia and the CIS run RuAPI in production.

Direct access from Russia

Acceleration nodes in several regions; for users in Russia, the nearest one is in Moscow. Tested from Moscow: the site and the API load in full, with no VPN.

Moscow nodeNo VPNLong replies arrive in full

Original models, no downgrade

Official upstreams only, no silent swaps to cheaper models. Genuine models — what you see is what you get, transparent pricing.

no-downgrade2–3 upstreamsauto failover

Transparent USDT / USDC payments

USDT / USDC in 5 minutes on TRC20, BEP20 or Polygon (USDT also on TON), gas under $1, fully on-chain. Balance never expires; unused funds are refundable within 7 days.

5 mingas < $1balance refund
# integrations

Connect in a minute

Enter the RuAPI URL and key in the tool you already use — no code changes. Each tool has a step-by-step guide.

Claude Code

Anthropic's official command-line coding assistant. No Anthropic subscription or account needed.

export ANTHROPIC_BASE_URL="https://www.ruapi.ai"
export ANTHROPIC_AUTH_TOKEN="sk-..."
claude
Step-by-step guide →

Cursor

Cursor stopped working in Russia in September 2026. The same models are available in Cline for VS Code:

Claude:       Anthropic, https://www.ruapi.ai
Other models: OpenAI Compatible, https://www.ruapi.ai/v1
API Key:      sk-...
Step-by-step guide →

Codex CLI

OpenAI's official command-line coding assistant; works with GPT models. Add to ~/.codex/config.toml:

model = "gpt-5.4"
model_provider = "ruapi"

[model_providers.ruapi]
name = "RuAPI"
base_url = "https://www.ruapi.ai/v1"
wire_api = "responses"
experimental_bearer_token = "sk-..."
model_catalog_url = "https://www.ruapi.ai/v1/models?for=codex"

# cheaper — replace with:
#   model = "gpt-6-luna"
Step-by-step guide →

Cline

An AI coding agent for VS Code. For Claude, pick the Anthropic provider — prompt caching makes it cheaper — then fill in:

API Provider:        Anthropic
API Key:             sk-...
Use custom base URL: https://www.ruapi.ai
Model:               claude-sonnet-5-5

# cheaper — replace with:
#   API Provider: OpenAI Compatible
#   Base URL:     https://www.ruapi.ai/v1
#   API Key:      sk-...
#   Model ID:     deepseek-flash
Step-by-step guide →

n8n

Workflow automation. Create an OpenAI credential (Credentials → New → OpenAI) and fill in:

API Key:  sk-...
Base URL: https://www.ruapi.ai/v1
Step-by-step guide →

OpenAI / Anthropic SDK

Your own code: change only base_url; the rest stays the same.

OpenAI SDK:    base_url = "https://www.ruapi.ai/v1"
Anthropic SDK: base_url = "https://www.ruapi.ai"
Step-by-step guide →
# providers

One KEY, every leading model

OpenAI and Anthropic protocol compatible. 80+ models supported.

OpenAI
OpenAI
21 models
gpt-6.2-astra · gpt-6.2-sol · gpt-image-2.6-flare
Anthropic
Anthropic
15 models
claude-fable-5-2 · claude-opus-5-5 · claude-sonnet-5-5
Gemini
Google Gemini
20 models
gemini-3.8-flash · gemini-3.1-pro-preview · gemini-3-pro-image-preview
DeepSeek
DeepSeek
4 models
deepseek-flash · deepseek-v4-pro
Grok
xAI Grok
1 model
Qwen
Alibaba Qwen
7 models
qwen4-max · qwen3.8-max
Zhipu
Zhipu GLM
3 models
glm-5.3 · glm-5.3-flash
Kimi
Moonshot Kimi
3 models
kimi-k3.1-pro · kimi-k3.1-pro-highspeed
Minimax
MiniMax
2 models
MiniMax-M3.1-Flash · MiniMax/MiniMax-M3
Doubao
ByteDance
4 models
doubao-seedance-2.5 · doubao-seedance-2.0
Make videos in the browser →

# Generated automatically from the pricing table, updated November 21, 2026 · full list →

# quickstart

5 minutes from $0 → sk-…

No card verification, no waiting list, no KYC.

01

Create account

$ rua register

Email OAuth · grab your sk-* key in 30 seconds

02

Top up

$ rua topup --usd

Auto-credited · no credit card, no SIM verification

03

Set BASE_URL

export OPENAI_BASE_URL=…

One env variable, every OpenAI / Anthropic SDK works · zero changes in your business code

# pricing

Original models, no swapping. Transparent pricing.

Per-token billing. No monthly fee, no minimum spend, no dynamic pricing.· live

modelinput $/Moutput $/M
claude-opus-5-5
$4.80
est. effective w/ cache ≈ $1.15
$24.00
claude-sonnet-5-5
$2.40
est. effective w/ cache ≈ $0.58
$12.00
gpt-5.5
$6.00
est. effective w/ cache ≈ $1.68
>272K: $12.00
$36.00
>272K: $54.00
gpt-5.4-mini
$0.90
est. effective w/ cache ≈ $0.25
$5.40
gemini-3.1-pro-preview
$2.40
est. effective w/ cache ≈ $0.67
>200K: $4.80
$14.40
>200K: $21.60
deepseek-flash
$0.17
est. effective w/ cache ≈ $0.0432
$1.44
deepseek-v4-pro
$1.58
est. effective w/ cache ≈ $0.36
$4.75
glm-5.3
$1.44
est. effective w/ cache ≈ $0.58
$5.02

Prices are per 1M tokens; the input column shows the regular (uncached) input price. The “est. effective w/ cache” line is an estimate at an 80% input-cache hit rate — it is not a discount, just a calculation for agent workloads. The price includes: payment in USDT/USDC without a foreign card, the native Anthropic API, one key for all models, support in Russian, and a refund of any unused balance within 7 days.

A grey line (e.g. “>200K”) is the model vendor’s official tier by prompt length: when a request’s prompt (input + cache reads and writes) exceeds that threshold, the whole request is billed at the higher tier. RuAPI adds no long-context surcharge of its own.

View full pricing table →

Claude API from Russia: pricing and payment → · ChatGPT API from Russia: pricing and payment → · DeepSeek API from Russia → · Gemini API from Russia →

# savings

Why it costs less in practice

Cached input is billed at a discount

For coding assistants and agents, repeated context (system prompt, open files, chat history) is often 80–97% of the input. Whatever hits the cache is billed at a discount: for Claude Sonnet 5, a cache read costs a tenth of the regular input price. On the pricing page, models that support caching show a small-print "est. effective w/ cache" figure — an estimate, not a discount.

cache 80–97%Est. effective w/ cache

A real workload: the same day, about 24× cheaper

One real day of Claude Code: 405M tokens. At RuAPI prices that's $200.69 on Claude Sonnet 4.6; the same tokens cost $8.39 on DeepSeek Flash and $44.52 on DeepSeek V4 Pro.

No hidden fees

You pay only for what you use: no monthly fee, no minimum spend, no long-context surcharge of our own and no dynamic pricing; the balance never expires. Each request's model, input and output tokens, and charge are shown in the usage log.

pay as you gobalance never expires
# compare

RuAPI vs. the alternatives

The common ways to use foreign LLM APIs from Russia.

RuAPIDirect (OpenAI / Anthropic)OpenRouterProxyAPI
PaymentUSDT / USDC or Visa / Mastercard, from $5Needs a foreign cardCard, USDC (USDT only via workarounds)Russian cards, SBP, rubles
Works from Russia without a VPNYesUsually needs a VPN (geo-restrictions)No, returns 403 (checked 9 Nov 2026)Yes
Models on one keyClaude, GPT, Gemini, DeepSeek, Qwen, GLM and more — 50+Only that provider's modelsMany providersMany providers
Native Anthropic API (Claude Code)Yes, works with DeepSeek tooAnthropic: yes, Claude onlyYes, but officially Anthropic models onlyYes, works with DeepSeek too
Model precisionOriginal models, no quantizationOriginal modelsSome providers serve FP8 / INT4—
SupportIn Russian, via TelegramIn EnglishIn EnglishIn Russian

Information about other services is taken from their websites (checked September 2026) and may change. Detailed comparisons: RuAPI vs OpenRouter · RuAPI vs ProxyAPI

# faq

Frequently asked

Yes, no VPN needed. The RuAPI site and API run through acceleration nodes in several regions; for users in Russia, the nearest node is in Moscow. We tested from Moscow: pages and long streamed API responses arrive in full.

Get started in 5 minutes

Sign up with Email. sk-* key in 30 seconds.

Create account