Evrona Chat: GPT, Claude, Gemini and 50+ models in one window

We've launched Evrona Chat — AI models right in your browser, no API key or code required. GPT, Claude, Gemini, DeepSeek, Grok, Qwen and dozens more models from our catalog live in one window. Replies are paid per token from the same balance as the API — no subscription, no separate plan. Open it at chat.evrona.ru or via the "Chat" button in the site header.

What is Evrona Chat?

Until now, the only way to try a model on Evrona was the API: create a key, plug it into an app or write some code. Great for developers, less so for everyone else. Now there's a regular chat, like ChatGPT or Claude, but with every chat model in our catalog to choose from.

The chat is part of Evrona, not a separate service. You sign in with your Evrona account and use the same balance — no extra registration.

Which models are available?

Every text model in the catalog works in the chat — 56 of them from ten developers right now:

OpenAI — GPT-6 Astra, GPT-5.6 Sol, Terra and Luna, GPT-5.5 and more;
Anthropic — Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5, Claude Haiku 4.5 and more;
Google — Gemini 4 Argon, Gemini 3.8 Flash, Gemini 3.1 Pro and more;
DeepSeek, xAI, Qwen, Moonshot, Z.ai, Xiaomi and MiniMax — DeepSeek V4 Pro, Grok 4.6, Qwen3.8 Max, Kimi K3.1 Pro, GLM-5.3 and more.

Pick a model from the list above the input box and switch it mid-conversation whenever you like. Each reply is labeled with the model that wrote it, so comparing models is easy. Models you starred in the catalog come first.

How much does it cost?

There's no subscription. Each reply is billed per token at the catalog prices and charged to your Evrona balance — the same one that pays for API requests. A short question to a fast model costs a fraction of a cent; long sessions with a large model cost more. If you don't use the chat, you don't pay for it.

What can it do?

Streaming replies. Text appears as it's generated — no waiting for the full answer.
Chat history. Your conversations are listed on the left; pick any of them up again from another device.
Assistants. Presets of a model, a system prompt and parameters, chosen before the first message.
Your own system prompt and parameters. Set temperature and max reply length per chat.
Remembers your choice. A new chat opens with the last model and parameters you picked.
Context meter. See how much of the model's context window is already used.
Copy and export. Copy a reply with one click or download the whole chat as Markdown (.md).

Enter sends a message, Shift+Enter adds a new line.

How is history stored?

Message text is stored encrypted on our servers. A chat is deleted automatically 30 days after its last message — or immediately, when you hit delete.

How do I start?

Open chat.evrona.ru or click "Chat" in the evrona.ru header.
Sign in to Evrona or create an account — it takes a minute. Your email needs to be verified.
If your balance is empty, top it up in the dashboard.
Pick a model, type a message and press Enter.

You can start typing before you sign in — your text stays in the box after registration.

Chat or API?

The chat is for hands-on work: asking questions, comparing models, writing text, reviewing code. The API is for software: Cursor, Claude Code, your own bots and services. They share one balance, so you can use both without separate top-ups.

FAQ

Do I need a subscription? No. You only pay for the replies you get, at catalog prices.

Do I need an API key? No. The chat works through your account; no key needed.

Does it work on my phone? Yes, the chat works in mobile browsers.

Are chat prices different from the API? No — the per-token price is the same.

Something's not working — where do I go? Contact support in your dashboard.

Summary

Evrona now has a chat: 50+ models in one browser window, pay-per-token from a shared balance, no subscription and no keys. Try it at chat.evrona.ru.