Simple credits. Clear data controls.
Core platform infrastructure is hosted in the EU. Paid plans start at €18/month.
Prices exclude VAT and any other applicable taxes.
Free
€0/mo
1,000 credits/month
Developers & open-source projects
- Image annotation tools
- AI-assisted labeling
- COCO export
- Organization for 1 member
- Up to 2 datasets
- Public cloud
The first Free organization receives 1,000 credits per month; additional Free organizations don't receive a monthly credit grant. Free organizations run on public cloud. Effectively anonymized and aggregated usage data may be used to improve Annota's models. User content is not used to train third-party models.
€0.01 flexible rate / credit
Pro
€18/mo
3,000 credits
Startups & privacy-first teams
- Everything in Free
- User content and organization data are never used to train our models
- Organization for up to 4 members
- Up to 40 datasets
€0.006 (−40% vs flex) / credit
Scale
€90/mo
20,000 credits
Mid-market AI teams
- Everything in Pro
- Organization for up to 6 members
- Up to 60 datasets
- SSO & audit logs (planned)
- Custom model fine-tuning (planned)
€0.0045 (−55% vs flex) / credit
Enterprise
Custom
Volume credits
Large companies & government
- Everything in Scale
- On-premise / air-gapped deployment
- Dedicated SLA
- Compliance support and custom controls
Custom volume pricing
Frequently asked
What's a credit?
Credits are Annota's internal currency: every AI-assisted operation on the platform draws from the same credit balance.
What uses credits?
Each Magic Select or Magic Annotate inference costs 1 credit. Batch annotation costs 1 credit for each asset and label processed. A COCO export costs 10 credits, and OCR costs 1 credit per processed page.
Can I buy extra credits?
Yes. Top-up credits never expire and start at €5, at a base rate of €0.01 per credit. Automatic discounts are 10% from €50, 20% from €250, and 30% from €1,000 for amounts below €2,000. Contact us for larger volumes.
Where is my data stored?
Customer content, databases, and primary storage are hosted in EU regions. Some supporting providers may process limited data internationally under GDPR safeguards. See Security for details.
Can you train on my data?
On Pro and above, User Content and organization data are not used to train Annota's models. On Free, effectively anonymized and aggregated usage data may be used for model improvement. User content is not used to train third-party models.
Do you offer on-premise?
Yes, on Enterprise, including air-gapped environments.
Get early access
Join the waitlist for early access and tell us how you plan to use Annota.
Thanks for joining! We'll be in touch soon.