Developer guide

Your first request.

From a workspace to a working request, using the tools you already know.

Create your workspace

Create an account and verify your email address. Add credits, then create an API key with a spending limit in your dashboard.

Install the SDK

pip install openai

Make it yours

Set RYNLER_API_KEY in your server environment and confirm the example model is available in the current catalogue. Generate a new idempotency identifier for each request.

Python
import os
import uuid
from openai import OpenAI

client = OpenAI(
    base_url="https://api.rynler.com/v1",
    api_key=os.environ["RYNLER_API_KEY"]
)

response = client.chat.completions.create(
    model="xiaomi/mimo-v2.6-flash",
    messages=[{"role": "user", "content": "Hello, world."}],
    max_tokens=1024,
    extra_headers={"Idempotency-Key": str(uuid.uuid4())}
)
print(response.choices[0].message.content)

Authentication

Send your key using Authorization: Bearer RYNLER_API_KEY. Keys are shown once, stored as hashes and can be revoked immediately. Keep them on your server, never in a browser bundle or a public repository.

List currently available models with GET /v1/models. Your key must include the inference scope.

Making requests

POST /v1/chat/completions checks estimated input plus requested maximum output against the selected model's verified context window. Read context_length, max_output_tokens and supports_tools from the model catalogue; null numeric limits mean the provider limit is unknown and platform safeguards apply. Supply an Idempotency-Key header containing 16–128 URL-safe characters.

Progressive streaming is available for models with a qualified streaming profile (on 2026-10-07: MiMo V2.6 Flash); other models deliver server-sent events after the full completion is validated. Function tools use tool_choice: "auto" on models advertising supports_tools: true. Preserve tool-call IDs, matching results and historical reasoning when continuing. The Responses API is not supported. Image input requires an enabled vision model and supported user-message image parts. Check GET /v1/vision-capabilities before use.

Errors & retries

CodeWhat to do
401Check your API key.
402Add credits or increase the key’s spending limit.
409Do not repeat a request with the same identifier. Inspect its status first.
429Reduce concurrency and retry later.
502The result or its token usage could not be verified. Review the reservation before retrying.
503The model or service is not currently available.

Security & data

Enable two-factor authentication in account settings. It is required for administrators. Browser sessions use HttpOnly cookies; account changes are protected against cross-site requests.

Rynler stores request metadata for usage and accounting, not the contents of your conversations. Inference services process prompts and outputs under their own data terms. Do not submit confidential or regulated data until those terms meet your requirements.

Read the current privacy and service information.