MEET ONE!

Developed for corporate needs, powered by AI and your data

Imagine you are building an AI application and 60% of what you need is already there!

ONE AI middleware is the ultimate hub to centrally provide all relevant GenAI functions for your company without having to develop or set them up from scratch, e.g. RAG (optimized for enterprise data), agents, chatbots, GPTs, complex prompt flows, connection of any number of data sources in near real time.

100% compliance - zero mis-use

Install ONE exactly where it is best for your company requirements: Microsoft Azure, Amazon AWS, Google Cloud Platform or on your own servers (on premise).

You have the same freedom with language models: OpenAI, Azure OpenAI, Amazon Bedrock, Google Gemini or self-hosted models - ONE supports them all (you can also use different models for different applications).

Some recent ONE live use cases

Automated
ESG Reporting

Implementation: 5 months

Costs: -67%
Workload: -81%

ROI: 9 months

Internal
Knowledge Database

Implementation: 1 month

Costs: -53%
Workload: -72%

ROI: 1 month

AI Accounting
Assistant

Implementation: 3 months

Costs: -59%
Workload: -69%

ROI: 4 months

Document Recognition
and Assignment

Implementation: 1 month

Costs: -85%
Arbeitslast: -92%

ROI: 2 months

Optimizing Procurement
with predictive AI

Implementation: 5 months

Production time: -9%
Savings: € 1.08 million p.a.

ROI: 3 months

Quotation preparation
by AI

Implementation: 1 month

Sales: +11%
Workload: -40%

ROI: 2 months

Specification
Agent

Implementation: 1 month

Sales: +7%
Costs: -82%

ROI: 1 month

Process Documentation
by AI

Implementation: 3 months

Workload: -85%
Costs: -34%

ROI: 4 months

Software Documentation
by AI

Implementation: 2 months

Workload: -90%
Costs: -25%

ROI: 1 month

Automated Test
Creation

Implementation: 2 weeks

Workload: -90%
Costs: -90%

ROI: 2 months

AI in Quality
Assurance

Implementation: 6 months

Workload: -30%
Product quality: +6%

ROI: 6 months

Automated Lead and
Customer Enrichment

Implementation: 2 months

Sales: +75%
Workload: -15%

ROI: 2 weeks

Lead Generation
by AI

Implementation: 3 weeks

Sales: +35%
Costs: -10%

ROI: 1 week

AI based
Research Agent

Implementation: 1 day

Workload: -65%

ROI: 1 month

Employee Onboarding
Agent

Implementation: 1 month

Workload: -15%
Onboarding: -1 week

ROI: 2 months

Customer Onboarding
Agent

Implementation: 1 month

Workload: -60%
Costs: -10%

ROI: 2 months

Customer
Support

Implementation: 4 months

Workload: -50%
Costs: -30%

ROI: 2 months

Webshop Shopping
Assistant

Implementation: 4 days

Sales: +10%
Shopping basket: +15%

ROI: 7 days

First level
Support Agent

Implementation: 5 days

Workload: -60%
Costs: -50%

ROI: 2 months

AI powered Application
Manangement

Implementation: 2 months

Workload: -80%

ROI: 1 month

Assessment with
AI

Implementation: 1 week

Workload: -95%
Costs: -70%

ROI: 1 month

Not sure which ONE version fits your needs best?

Micro

For solopreneurs and small use cases

29 €*

per month, billed monthly

1User

Email support
Hosted in Germany
Chatbots

no creditcard required

Professional

Most important features

Users	The number of users for your ONE instance is defined by your chosen plan. Every chatbot requires one user license.
Bring your own LLM	ONE provides a basic setup, which is running on GPT-4o (hosted in Sweden). For compliance reasons we recommend you to connect your own LLM instances. The number of models that you can connect is not limited, regardless of your chosen plan.
Indexes and data sources	You can connect as many data sources as you like. For non-realtime data sources (e.g. file storages) there is a fair use policy, which adds a rate limiting on very large data sources, so that index creation is not blocking other users from an outstanding experience on using Nexus. Keep in mind, that large data sources take longer for (re-)indexing and also slow down your RAG results. Apart from that, we are working on a fair-use model, meaning, if your data sources would block other users from reasonable performance, we reserve the right to lower the re-indexing rate for these data sources.
Large Language Models (LLM)	We are continuously working on extending the list of supported Language Models. Generally spoken we are supporting all OpenAI, Microsoft Azure, AWS Bedrock, Mistral and Anthropic models out of the box. Additionally, we are extending the supported models from Huggingface nearly every month. Let us know, if you are missing a model, that would help you.
Connectivity	ONE can easily be integrated into any business and software solutions via its REST and Stream APIs.
Web UI and API-first	ONE comes with a solid WebUI for direct use cases. As it is built as an API-first solution, integration in any kind of product is easily possible.