Free Variant

Not yet available.

Some models on OpenDunes have a free tier served by providers that make capacity available at no cost. Append :free to any eligible model slug to explicitly request the free tier. No DA balance is deducted for free-variant requests.

Usage#

curl https://opendunes.com/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENDUNES_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meta-llama/llama-4-maverick:free",
    "messages": [{ "role": "user", "content": "What is the capital of Algeria?" }]
  }'

The :free suffix is part of the model slug — pass it as-is in the model field. The response model field echoes the slug with the suffix, so you can always tell which tier handled the request.

How it works#

Free tiers are served by providers that offer limited capacity at no charge. OpenDunes routes :free requests exclusively to those providers. Because capacity is finite, free-tier requests are subject to:

• Lower rate limits than paid tiers — typically 20 requests/minute per key (exact limits shown in your dashboard).
• Higher queuing probability during peak hours — free requests may wait longer before being processed.
• Reduced availability guarantees — if the free-tier provider is at capacity, the request fails rather than falling back to a paid provider.

To get guaranteed throughput, use the model without the :free suffix, or combine it with model fallbacks.

Discovering free models#

Filter the public model catalog to show only free-tier eligible models:

GET /api/models?free=true

The response includes each model's :free slug and any capacity notes. You can also browse them at /models?free=true.

Pricing (DA)#

Zero. Free-variant requests are not billed and do not draw down your DA balance. The X-Balance-Available response header is unchanged after a :free request.

Rate limiting still applies — see the X-RateLimit-* headers in every response.

Using with the OpenAI SDK#

import os
from openai import OpenAI

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

resp = client.chat.completions.create(
    model="meta-llama/llama-4-maverick:free",
    messages=[{"role": "user", "content": "Translate 'good morning' to Darija."}],
)
print(resp.choices[0].message.content)

Note. The :free suffix only works on models that have a free tier. Appending it to a model with no free tier returns a 404 model_not_found error. Check the catalog to confirm availability.