# Phoeniqs Chat – Technical Overview

# Overview

The Phoeniqs Chat is an enterprise-grade conversational AI platform designed for secure deployment within trusted environments. It supports browser-based, hybrid, and fully on-premises use cases and is purpose-built for internal organizational functions such as HR helpdesks, IT support, developer enablement, and compliance Q&A.

The platform provides secure access to a curated set of open-source large language models, hosted and executed entirely within Phoeniqscontrolled environments. Models are selected based on workload characteristics and can be accessed consistently through the platform’s conversational interface or APIs.

# 💬 Enterprise Conversational Interface (LibreChat-based)

The Phoeniqs AI Platform includes a LibreChat-based conversational interface that enables secure, intuitive interaction with Phoeniqsmanaged AI services.

Key capabilities include:

  • Chat-based interaction with Phoeniqshosted conversational AI agents
  • Local session history with user-controlled retention, export, and clearing
  • Session-scoped document uploads (up to three PDFs) for Mini-RAG contextual grounding
  • In-browser and on-prem UI deployment without reliance on external SaaS frontends
  • Consistent interface access across supported LLMs, independent of model choice
  • Integrated with Phoeniqs LLM offering (MaaS) - Model Selection by Use Case? More info here

The conversational interface serves as the interaction layer only; all model execution, data access, and security controls are managed by the Phoeniqs AI Platform runtime.

# 🚀 Lightweight, Flexible Deployment

Supports rapid adoption through browser-based access while enabling hybrid and fully on-prem deployments.

# 🔧 Open-Source Model Backbone

Built on open-source LLMs selected for transparency, adaptability, and performance, with support for customization and future in-session model switching.

# 🧪 Trial and Production Access

Provides trial and production tiers, including:

  • Serverless endpoints for model execution
  • Model Execution & Indexing Units (MEIUs), including Retrieval-Augmented Generation (RAG)
  • Predictable consumption models for experimentation and scale

# Accessing the Platform

To access the Phoeniqs AI Platform:

  • Create or receive a user account
  • Select an access tier: Trial or Production
  • Log in to the platfo

For questions or support, contact the technical support team.