Skip to content
Regolo Logo
Regolo Logo
● Official Announcement Effective November 1, 2026

Regolo pricing update

We are updating Pay As You Go rates and subscription model multipliers.

LIVE NOVEMBER 1 We updated our rates to reflect model capabilities, allocate dedicated compute hardware, and deliver higher throughput. Several models become more affordable, while other models reflect their true inference requirements.
Current pricing rates and plan rules remain active until that date.
Explore New Rates ↓ Ask on Discord

Pay As You Go pricing

All listed rates show the cost in euros for each one million processed tokens.

Model Input Output
gpt-oss-20b €0.10 €0.20
qwen3.5-9b €0.10 €0.20
brick-complexity-pro Semantic Smart Router €0.10 €0.20
gpt-oss-120b €0.20 €0.70
mistral-small-4-119b €0.20 €0.70
gemma4-31b €0.40 €0.60
qwen3.8-27b €0.50 €3.40
qwen3.5-122b €0.60 €3.40
apertus-70b €1.00 €3.40
glm5.2 €2.30 €6.90

Subscription model multipliers

The same model multipliers apply across all subscription plans, they determine how usage counts against your included token allowance, not a separate price in euros.

Model Usage multiplier
glm5.2 5×
qwen3.5-122b 3×
apertus-70b 2×
qwen3.8-27b 2×
gemma4-31b 1×
mistral-small-4-119b 1×
gpt-oss-120b 1×
brick-complexity-pro 0.5×
qwen3.5-9b 0.5×
gpt-oss-20b 0.5×
For example, one million applicable tokens processed with a 0.5× model consume 500,000 tokens of your allowance. If you use a 2× model, one million tokens consume 2,000,000 tokens of your allowance.

Go further with Brick

Up to 80% Inference Cost Savings

Brick routes requests to suitable models based on task requirements, we published benchmarks to show inference cost savings of up to 80% compared with using the highest-cost model in the benchmark pool for every request.

Without Brick (Uniform fixed model) 100% Cost Baseline
With Brick (Dynamic complexity routing) 20% Cost Up to -80% Saved

These are allowance comparisons, not guaranteed workload gains. Actual savings and usable capacity depend on your workload, token volumes, and plan limits.

OpenAI compatible: switch with 1 line of code

You do not need to rewrite your software or change your libraries, change the base URL to connect your applications to European hosted models immediately with zero data retention.

python · openai-sdk-drop-in.py 
from openai import OpenAI

# 1 Line Change: Standard OpenAI SDK with Base URL
client = OpenAI(
    base_url="https://api.regolo.ai/v1",
    api_key=os.environ.get("REGOLO_API_KEY")
)

response = client.chat.completions.create(
    model="brick-complexity-pro",  # or glm5.2, qwen3.8-27b, mistral-small-4-119b
    messages=[{"role": "user", "content": "Execute workflow analysis..."}]
)

Frequently asked questions

Find answers below about the updated rate schedule, the transition period, and subscription allowances.

Why are the rates changing?
We updated the rates to show exact costs for input tokens and output tokens. Prices now reflect true inference requirements and allocate dedicated graphics processor capacity to deliver higher throughput.
When do the changes take effect?
Current pricing rates and plan rules remain active until that date.
How do subscription multipliers work?
Multipliers determine how your usage counts against your included token allowance without a separate fee in euros. For example, a 0.5× model uses 500,000 allowance tokens when you process one million tokens.

Questions about your plan?

If you have questions about how these rate updates affect your plan, you can talk with the Regolo technical team on Discord.

Certified European Infrastructure & Sovereignty
ISO/IEC 27001
ACN Qualified
Zero Data Retention
CISPE Code of Conduct
100% Green Energy