kosareva.cloud
APIOpenAI- and Anthropic-compatible

One AI API
for all your models

One API key for OpenAI, Anthropic, Google, xAI, DeepSeek and others. Your SDKs and tools stay the same: you change the base URL and the key, not the code.

The model catalog is in Russian for now; prices are in rubles.

OpenAI formatbase_url = https://api.kosareva.cloud/v1
Anthropic formatANTHROPIC_BASE_URL = https://api.kosareva.cloud
/ One key

100+ models behind one key

Text, code, images, video and speech from different vendors — with one balance, one bill and one place to see where the money went.

Switch vendors with one parameter

Claude, GPT, Gemini, Grok, DeepSeek, Qwen, Kimi, GLM, MiniMax and more. Change the model name in the request and you are working with another vendor — the rest of the code stays as it is.

modelclaude-sonnet-5-5modelgpt-6-1-solmodelgemini-3-1-pro

A key for every project

Create separate keys for projects and teammates, each with its own spending limit. You always know which project spent what.

Usage you can actually read

Spending by key and by model, down to a single request. Alerts when a key reaches its limit, and a report you can export for your accountant.

Images, video and speech too

The same key generates images — GPT Image 2, Nano Banana 2, FLUX — and video with sound — Veo 3.1, Kling 3.0. Video runs asynchronously: you submit a task and pick up the finished file. Text-to-speech is in the catalog as well.

/ Quick start

Your first request in three steps

  1. /1

    Sign up

    Create an account in the dashboard with your email.

  2. /2

    Top up the balance

    From 50 ₽ with a Russian bank card or SBP. Companies can pay by invoice — see payment.

  3. /3

    Send a request

    Create an API key, pick a model and point your SDK at the OpenAI-compatible endpoint.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_KOSAREVA_KEY",
    base_url="https://api.kosareva.cloud/v1",
)

resp = client.chat.completions.create(
    model="claude-sonnet-5-5",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)
/ Tools

Works in your editors and agents

Any client that speaks the OpenAI or Anthropic API works with the same key.

VS Code
Code editor

Our extension from the Marketplace brings the models into the built-in chat and agent mode. No Copilot or GitHub account needed.

Claude Code
Terminal agent

Set ANTHROPIC_BASE_URL and your key, and Claude Code works through the Anthropic-compatible endpoint.

Cursor
Code editor

An AI editor built on VS Code. Connects by overriding the base URL.

OpenClaw
Messengers

An AI agent for Telegram, Discord and Slack that runs on your own server.

Also works with
  • Codex CLI
  • Continue
  • Cline
  • Aider
  • Hermes
  • n8n
  • LangChain
  • LlamaIndex
  • Vercel AI SDK
  • LiteLLM
Setup guides (in Russian) →
/ Pricing and payment

Prices in rubles, pay as you go

Every model has a fixed price in rubles. There is no subscription and no monthly fee: you top up the balance and pay only for the tokens you use. What is left on the balance does not expire.

The model catalog

Prices per million tokens, per image and per video for every model. The catalog is in Russian for now, but model names and numbers read the same in any language.

Open the catalog →

How to pay today

  • ✓Individuals: a card issued by a Russian bank (Mir, Visa, Mastercard) or SBP, Russia’s Faster Payments System. From 50 ₽, right in the dashboard.
  • ✓Companies: bank transfer against an invoice from Kosareva Cloud LLC, Republic of Armenia, from 1,000 ₽. We sign a contract and provide closing documents: invoice and act.

Paying from abroad

We don’t accept cards issued outside Russia yet. If that is what stands between you and the API, write to us at ceo@kosareva.cloud or in Telegram @kosareva_cloud and tell us how you would like to pay.

Not translated yet: the model catalog, setup guides, articles and legal documents (public offer, privacy policy, SLA).

Start with one key
and the models you need

Sign up, top up from 50 ₽ and send your first request. Questions before you start? Write to us directly.