Choose the plan that fits your needs
Zero Data Retention, Fast and Secure AI
Start free for a month, subscribe for more capacity, or go flexible with pay-as-you-go.
Access to all features. Scale when you need. Pay only for what you use.
START YOUR 30-DAY FREE TRIAL
Get full access to all Core Models. No credit card required, no commitment. Sign up and start building.
Start Free Trial1 month duration
Full access for 30 days, then choose a plan
1M tokens per day
Daily token limit to explore all features
Stricter rate limits
Fair usage throttling applies
All Core Models
Complete access to chat, embedding, and more
Regolo Core Models
Best-in-class model performance, effortless autoscaling, and blazing fast cold starts mean you get the most out of each GPU, saving money along the way.
Models library pricing
Here you can explore the full list of supported models, together with their token pricing* and limits. Use this table to plan your workloads and choose the most cost‑efficient models for each task.
Large Language Models
Text and multimodal chat models priced per million tokens (input and output).
Enterprise pricing may apply for high-scale or custom deployments.
Talk with our engineers to fit a special offer for your custom needs.
*Prices exclude VAT, applied only in Italy
How do included tokens work in our plans?
Every plan includes a daily allowance of tokens you can spend on any Core chat model. Not all models consume it at the same speed: each one carries a usage multiplier that decides how much of your allowance every request actually uses. Choose the right model for each task, and the same plan goes a lot further.
How it works
Tokens count toward your daily allowance, multiplied by the model's factor.
Multipliers work in both directions: compact models (below ×1) consume less allowance than their raw token count, so your plan lasts longer.
Every chat model is included in every plan, from Core to Scale: multipliers only change how long your daily allowance lasts, never which models you can use.
Think of it as a dial you control: pair compact models with routine work and save frontier ones for the moments that truly need them. Users who balance the two consistently get more total work out of the same plan: the allowance becomes a budget you optimize, not a ceiling you hit.
One example, three models
Same 10,000 tokens, three very different impacts on your daily allowance.
In other words: with the most efficient models your plan can deliver up to 2× more tokens than its face value. That is real headroom you unlock simply by picking the right model for each task.
Chat model multipliers
* Consumption projection based on the Boost plan (50M tokens per day). It assumes the selected model is the only one you use all day.
Deploy models from Huggingface
Only pay for the compute you use, down to the minute. Deploy using Custom Models.