1NCE AI: EU-Hosted Claude + Multi-Model AI Platform
Run 17 AI Models Through One API
///
)
Powered by AWS and models from leading providers
)
){kind=logo}
){kind=logo}
){kind=logo}
1NCE AI platform gives your team access to 17 models, including Claude, Llama, Mistral and more, through a single OpenAI-compatible endpoint, at the same price you'd pay AWS directly. Your first API call takes under 3 minutes.
1NCE AI Insights:
17+ hosted models
99.9% target availability
< 3 min to first API call
One single API Key
No AWS account required
5€ free credits to start
A Smarter Way to Run AI at Scale
Built for IoT teams, enterprise developers, and platform engineers.
One API Key across Every Model
Access all available models through a single key, without separate provider accounts or IAM configuration.
OpenAI-Compatible API
Use your existing OpenAI SDKs and workflows. Point your application to the 1NCE AI endpoint and switch models with a single parameter.
Governance and Spend Control
Set rate limits and spending caps across teams and models. Monitor usage with audit logs and keep AI costs under control.
IoT-Native Integration
Already using 1NCE connectivity? Add AI through the same portal, on the same invoice, with support from the same team.
EU Data Residency for Claude
Claude runs in AWS Germany (eu-central-1), keeping prompts and outputs within the EEA. Other models run outside the EEA and are clearly labeled.
Zero Data Training Required
Never used to train or fine-tune models; contractually committed for Bedrock-hosted models (incl. Claude); US-routed models follow their provider's terms.
Unified Billing and Spend Tracking
Get one invoice for all usage. Monitor consumption in real time, manage prepaid credits, and control budgets across teams.
An AI Platform Created for Responsible AI
A minimal, production-ready platform designed to let you build.
Multi-Model Routing
Automatically route requests to an alternative model if the primary model becomes unavailable.
Usage Dashboard
Track usage by model, API key, and team. Monitor costs in real time and download detailed invoices.
Self-Service Portal
Create accounts, generate API keys, configure budgets, and manage access from one AI platform, without waiting for a sales process.
Full Audit Logging
Every API call includes model information, token counts, latency metrics, and costs.
)
How It Works
Live in Minutes
Create Your Account
Sign up with your work email. Existing 1NCE customers can use their current credentials.
Generate an API Key
Choose your models, define optional spending limits, and create a scoped API key.
Make Your First Call
Point your OpenAI-compatible client at api.1nce.ai/v1. Drop-in compatible, without new SDKs or refactoring.
Track Usage and Billing
View token usage by model and API key. Export reports and receive transparent billing.
Pricing
Pay Only for What You Use
Usage-based pricing means you only pay for consumed tokens at standard AWS Bedrock rates. Skip minimum spend or monthly fees. Start with 5€ free credits and top up at any time using Stripe or bank transfer for larger accounts.
What Teams Build with 1NCE AI Platform
AI Features Inside Your Product
Add AI capabilities to your SaaS platform without managing multiple model providers and accounts.
IoT Data Interpretation
Turn raw telemetry into plain-language reports, anomaly summaries, and operational insights.
Automated Document Processing
Summarise contracts, extract structured data, and review compliance documents with a single API call.
Internal AI Chat for Regulated Organisations
Deploy OpenWebUI or LibreChat on your own infrastructure while keeping conversations, audit logs, and user controls inside the EEA.
Get Started with 1NCE AI Now
Deploy AI with confidence through EU data residency, enterprise-grade compliance, one API for every model.
FAQ
Where exactly is my data processed? Is "EU-hosted" the same as GDPR-compliant?
Where exactly is my data processed? Is "EU-hosted" the same as GDPR-compliant?
It depends on the model. Claude models are processed in AWS data centres in Germany (eu-central-1), so Claude prompts and outputs stay within the European Economic Area for the full request lifecycle. Llama, Mistral and US-routed models (for example, GPT) are processed in their provider regions outside the EEA. Every model’s processing region is labelled, so you can restrict your team to EU-hosted models where data residency is a hard requirement.
EU data residency is a necessary condition for GDPR compliance, but not the only one. 1NCE AI also provides a signed Data Processing Agreement (DPA) and a Technical and Organisational Measures (TOM) document: the two artefacts your DPO will ask for. Both are available on request at sign-up, not weeks into a procurement process.
Can I get a signed DPA before we go to production?
Can I get a signed DPA before we go to production?
Yes, and you don't have to wait to get into production first. The DPA, TOM document, and sub-processor list are available on request as soon as you create an account. Contact the sales team or request them directly from the portal.
These are standard documents, not custom negotiations. For most enterprise procurement workflows, they're ready to pass to legal on the same day you sign up.
Will my prompts or outputs be used to train AI models?
Will my prompts or outputs be used to train AI models?
No. Prompts and model outputs are not used to train or fine-tune foundation models. For models served through AWS Bedrock (including Claude), this is a contractual commitment in the AWS Bedrock service agreement and applies to every Bedrock request. US-routed models such as GPT are served through their own provider (OpenAI) and are covered by that provider’s API terms; we surface the applicable terms per model so your legal team can confirm before use.
For Bedrock-served models, this guarantee is baked into the AWS Bedrock service agreement, not a policy that can be changed by a product update. If your legal team wants the specific clause, it’s available in the AWS service terms and referenced in the 1NCE DPA.
We already have an AWS account and use Bedrock. Why would we use 1NCE AI instead?
We already have an AWS account and use Bedrock. Why would we use 1NCE AI instead?
Direct Bedrock access requires separate per-model access requests (each taking days), IAM policies per model, and no built-in team spend controls or audit dashboard. You also manage billing across however many models you use.
1NCE AI gives you all 17 models on day one, a unified spend dashboard, per-key rate limits, at the same price as Bedrock. You also get the 1NCE governance layer on top: audit logs, PII redaction, and role-based key management, none of which Bedrock provides natively.
We're already a 1NCE connectivity customer. Do we need a separate account?
We're already a 1NCE connectivity customer. Do we need a separate account?
No. 1NCE AI is accessible from the same portal you already use for connectivity. Sign in with your existing credentials, generate an API key, and AI usage appears on the same invoice as your connectivity services. There's no separate onboarding, no new account, and no separate support contact.
What happens if we exceed our prepaid credit balance?
What happens if we exceed our prepaid credit balance?
Service stops when the balance reaches zero, it doesn't continue and bill you retroactively. You can set per-key spend caps to stop individual keys before the organisation-level balance runs out, and the dashboard shows real-time consumption so you can top up before it's an issue.
For accounts above a defined monthly spend threshold, bank transfer is available as an alternative to card top-ups.
What models are available, and are new models added automatically?
What models are available, and are new models added automatically?
The current lineup includes Claude (EU-hosted), Llama, Mistral, Amazon Nova, and others (US regions), 17+ models in total. New models are added as they become available on AWS Bedrock EU region, and you access them through the same API key without any additional configuration or access requests.
Is PII redaction enabled by default, or do we have to configure it?
Is PII redaction enabled by default, or do we have to configure it?
PII redaction is optional and configured per API key: it is not enabled by default. You switch it on for the keys where it's needed, without any changes to your application code. This is useful for teams processing documents or user input that may contain personal data, particularly where you want a second layer of protection on top of EU data residency.
Newsletter