Solution Overview
XR5.0 LLM Engine is an AI-powered technical assistance platform designed specifically for enhancing extended reality and other applications with customised LLM agents. It combines advanced language models with your organisation’s technical documentation to provide accurate, context-aware answers to your questions.
The platform allows you to interact with customizable AI agents implementing Retrieval Augmented Generation (RAG) on your technical documents, manuals, and specifications to help you find information quickly and efficiently.
1. Request Access
To create an account, please request access by contacting [email protected]. Once approved, you will be able to register and set your password using your provided email.
2. Sign In
Use your email and password to access the dashboard. Once logged in, you’ll have access to all features including document management, agent customization, and the chat interface.
3. Explore the Dashboard
From the dashboard, you can view your agents, manage documents, and access all platform features. The sidebar navigation provides quick access to all sections.
The Agent Testing page is where you interact with your AI agents. Here’s how to get the best results:
Asking Questions
– Be specific: “What are the technical specifications for [product/feature]?” instead of generic questions
– Provide context: Mention specific models, products, or situations
– Ask follow-up questions to dig deeper into topics
– The agent remembers your conversation history within each session
Tips for Better Results
– Make sure relevant documents are uploaded to the agent’s knowledge
– Use clear, technical language when appropriate
– If you don’t get a good answer, try rephrasing your question
– The agent can only answer based on the documents it has access to
You can integrate XR5.0 LLM Engine into your own applications using our REST API. There are two ways to authenticate:
Authentication Methods
Option 1: API Tokens (Recommended)
Create long-lived API tokens in the Settings page. These tokens are secure, include your agent and namespace information, and don’t require you to share your password with applications.
curl -X POST https://chat-api.xr50.eu/api/chat \
-H “Authorization: Bearer YOUR_API_TOKEN” \
-H “Content-Type: application/json” \
-d ‘{“query”: “What are the key features?”}’
Benefits: Tokens can be revoked without changing your password, have configurable expiration dates, and include embedded agent access information.
Option 2: Login Token
Obtain a temporary access token by logging in with your credentials:
curl -X POST https://chat-api.xr50.eu/api/auth/token \
-H “Content-Type: application/x-www-form-urlencoded” \
-d “[email protected]&password=your-password”
This returns a temporary access_token token that expires after a short period.
Use this token in the same way as an API token.
Querying AI Agents
Send questions to your agent using the chat endpoint:
curl -X POST https://chat-api.xr50.eu/api/chat \
-H “Authorization: Bearer YOUR_TOKEN” \
-H “Content-Type: application/json” \
-d ‘{“query”: “What are the key features of this system?”}’
The response includes the AI-generated answer along with relevant source documents from the knowledge base.
Uploading Documents
Add documents to your agent’s knowledge base programmatically:
curl -X POST https://chat-api.xr50.eu/api/upload \
-H “Authorization: Bearer YOUR_TOKEN” \
-F “[email protected]” \
-F “pilot=pilot-3” \
-F “doc_type=Manual”
Supported formats: PDF, DOCX, and TXT files. Documents are automatically processed and indexed.
Complete API Documentation
For detailed API documentation with all available endpoints, request/response schemas, and interactive examples, visit https://chat-api.xr50.eu/docs. The interactive documentation allows you to test endpoints directly from your
browser.
© 2025 Innov-Acts Ltd.
Supported by the HEU XR5.0 project. XR5.0 has received funding from the European Union’s research and innovation programme under grant agreement No. 101135209.