Access without a VPN
One platform for AI models
Transparent pricing in rubles, one balance, and documents for companies.
POST /v1/chat/completions
Authorization: Bearer $PROVOD_API_KEY
Content-Type: application/json
{
"model": "openai/gpt-5.4",
"messages": [{ "role": "user", "content": "Hello" }]
}Popular models
The public catalog shows routeable model families, identifiers, capabilities, and ruble prices. Availability follows the current catalog.
Connect with a familiar SDK
Use the OpenAI SDK with the provod.ai base URL and an explicit public model identifier.
- Streaming responses
- API-key spend limits
- Request-level usage
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.PROVOD_API_KEY,
baseURL: "https://api.provod.ai/v1"
});
await client.chat.completions.create({
model: "openai/gpt-5.4",
messages: [{ role: "user", content: "Hello" }]
});Chat for working with models
Discuss tasks, attach documents, and switch models in one workspace.
Try chat
Billing and control
Top up your ruble balance and see spending in one place.
- API key limits
- Request and charge history
- Invoices and closing documents for your company

Start in four actual steps
01Create an account
One account for API, chat, and team workflows.
02Add funds
Fund the balance in rubles before starting work.
03Create an API key
A dedicated key connects the application to the API.
04Send a request
The first request verifies the key, endpoint, and model.
Helpful answers before you start
Short guidance for common account, billing, and model questions.
Create an account, add funds, create an API key, then send a first request with a model from the catalog.
No. Add funds to the balance and pay for confirmed usage at the current published rate.
Some models in a personal workspace can require a first real top-up. Check the current catalog for the unavailable reason.
Sign in using the method linked to your account, then check your workspace, balance, and keys.
Recent guides
Explore API setup, models, usage controls, and team workflows in the blog.

Async-вызовы и Batch API в LLM: как сэкономить до 50% и ускорить обработку
Performance-гайд 2026: async LLM-вызовы на Python (asyncio + aiohttp), Batch API у OpenAI и Anthropic со скидкой 50% на input/output, расчёт savings, очереди задач, retry-паттерны, когда брать async, когда batch. Точные числа throughput и стоимости через единый шлюз provod.ai.

B2B чек-лист: 12 вопросов поставщику LLM API перед подписанием договора
B2B чек-лист по выбору поставщика llm api: 12 вопросов в четырёх блоках — Compliance, Pricing, SLA, Security. Договор chatgpt юрлица, закрывающие документы через ЭДО, наценка на токены, SLA на доступность, изоляция ключей, политика по данным. С разбором правильных и опасных ответов.

Embeddings и векторный поиск: полный RAG-стек 2026 для русскоязычных проектов
Глубокий разбор RAG-стека 2026 для русскоязычных проектов: модели embeddings (OpenAI text-embedding-3, Cohere v3, Voyage 3), векторные БД (pgvector, Qdrant, Chroma), стратегии chunking, hybrid search и retrieval evaluation. С реальными цифрами стоимости и Python-кодом через единый шлюз provod.ai.