CRM / SUPPORT
Intelligent chatbots
Add a chatbot to your site or CRM. Send a question with customer context, get a relevant reply, and route complex cases to an agent.
Faster support handling
FOR DEVELOPERS AND IT TEAMS
One API.
AI in your product.
Connect GPT, Claude, Gemini and free models to your services. A familiar request format, one key and a shared balance.
OpenAI SDK compatible · JSON and streaming responses
olesa / api-gateway
REST + SSETHE REQUEST JOURNEY
Your service
client
O·LESA API
gateway
GPT-OSS 20B
inference
Ready to send a request
01POST /v1/chat/completions
02Authorization: Bearer YOUR_API_KEY
03model: gpt-oss-20b-free
GPT / Claude / Gemini+ GPT-OSS
INTERACTIVE DEMO
See a request in action.
Choose a prepared scenario and watch the request flow through the API. A sample text or image response appears at the end.
playground.ts
INTERACTIVE DEMOYour service
client
O·LESA API
gateway
GPT-OSS 20B
inference
Ready to send a request
02 / responseINTERACTIVE DEMO
A sample response will appear here
Choose a scenario and run the demonstration.
This scripted demonstration uses prepared requests and responses; it does not call a model. The image illustrates a result, not a tested image API endpoint. Create an API key for real requests.
01
A familiar interface
OpenAI-compatible Chat Completions: messages, JSON replies or an SSE stream. Works with Python, JavaScript and HTTP integrations.
02
A model for each task
Compare quality and cost. Change the model field in your request to switch between available models.
03
Access and spending controls
Create and revoke keys in settings. Available models use a shared balance; costs depend on the model and token usage.
INTEGRATION
From a test to your code.
Use the OpenAI SDK or an HTTP request. Set the O·LESA base URL, your key and a model ID from the catalog.
DocumentationPythonNode.jsHTTPSSE
HTTP / cURL
curl 'https://api.o-lesa.com/v1/chat/completions' \
-H "Authorization: Bearer $OLESA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-oss-20b-free",
"messages": [
{
"role": "user",
"content": "Write a short reply to a customer asking whether they can connect AI to their CRM through an API. Explain the benefit and suggest a next step. Up to 3 sentences."
}
]
}'Server-side example. Replace OLESA_API_KEY with your key; the full value is shown once when you create it.
BUILD INTO YOUR WORKFLOW
Less routine work. More product possibilities.
CMS / E-COMMERCE
Content for new pages
Generate product descriptions, meta tags and text variants. Return results to your CMS through your integration.
Faster content production
INTERNAL TOOLS
Data and document analysis
Send report or log text, request findings in a useful format and pass them to the next step of your workflow.
Less manual preparation
A KEY IN YOUR ACCOUNT
Three steps to your first application request.
01
Create an account
Sign in with Yandex, Google or an email code. Available methods depend on the domain.
02
Set up an API key
API settings open right after sign-in. Create a key with model access and save it.
03
Connect your service
Store the key in your server environment, choose a model and send a request.
Before you connect
Do I need an account for the demo?
No. This page shows prepared requests and responses; it does not call a live model. Requests from your application need an account and a personal API key.
What does the model API support?
Text Chat Completions in an OpenAI-compatible format: standard JSON responses and SSE streaming. Find available model IDs in the catalog and documentation.
How are requests billed?
Paid models charge used tokens to your shared balance. Free models do not charge your balance but share provider limits and may be temporarily unavailable.
Where can I manage my key?
In Settings → API. Create a key with the permissions you need or revoke it. The full key is available only when created.
O·LESA / API
Your next request comes from your application.
Create an account, get a key and connect models to your service.