API capabilities
and connection terms
OpenAI-compatible
Works with existing SDKs and code: only the base URL and the key change.
Streaming responses and semantic search
The reply streams word by word as it is ready; there are models for semantic search and tool calling.
Developer console
Keys, limits, request logs and usage statistics.
Access from Russia without a VPN
The request goes to our address directly; no workarounds for the connection are needed, nor a foreign card.
Billing in rubles
1 credit = 1 ₽, you pay only for the tokens you use.
How to start
- 1Create an account and an API key in the console
- 2Replace the base URL with https://api.yenisei.ru/v1
- 3Pick a model in the catalog and send the first request
Integration with popular tools
Names and logos belong to their owners. Yenisei is not affiliated with these projects.
Try it without leaving the page
A live request to our models: the same API you will get.
Ask something or pick an example below.
{ "model": "yenisei-100", "messages": [ { "role": "user", "content": "Reply to a customer's review about a late delivery" } ], "stream": true, "temperature": 1, "top_p": 1}The text you enter is processed to show the model's reply. Do not enter personal data here, neither yours nor anyone else's. Details are in the personal data policy.
Whose models are in the catalog
Yenisei's own models and open models from other makers.
yenisei-100, yenisei-code, yenisei-max, …
mistral-nemo, mistral-small-3.2, voxtral-small
qwen-flash, qwen3.8-27b, qwen3-embedding-8b, …
gpt-oss-20b, gpt-oss-120b
granite-8b
nemotron-nano-30b
deepseek-flash, deepseek-chat, deepseek-pro, …
phi-4
llama-4-scout, llama-4-maverick
kimi-k2-thinking, kimi-k2.6, kimi-k3, …
bge-m3
mimo-v2.5, mimo-v2.5-pro
glm-4.7-flash, glm-5.2, glm-5.3, …
command-a
gemma-4-31b, gemma-4-26b
ling-3.0-flash, inkling
FAQ
Try Yenisei API
Sign in with your email and the code from the letter. The Basic plan is free.