LLMUI on-premise AI plateform

LLMUI provides an on-premise platform for managing and orchestrating language models, suitable for various enterprise needs. Key capabilities include: • Two versions: Core (open-source) and Enterprise (proprietary) • Andy multi-role agent for complex task orchestration • Multi-instance Ollama management • Git connectors and Proxmox integration • Mobile PWA and push notifications for accessibility • Comprehensive enterprise security features LLMUI Core v0.5.0 offers a simple, efficient open-source solution, ideal for developers and small teams looking for basic model management. It supports default models like phi3:3.8b and gemma2:2b for workers, and granite4:micro-h for merging, all running locally with an Nginx proxy and Ollama backend. This version provides a foundational toolkit for local language model deployment without extensive setup. The Enterprise v4.3 version elevates capabilities with the central 'Andy' agent, designed to execute messages through a configurable multi-role pipeline including task analysis, tool management, parallel workers, and intelligent merging. Andy features a ReAct loop for tool orchestration, streaming SSE, and simultaneous parallel conversations. It supports execution on remote hosts via SSH for both Windows and Linux, personal notes, and push notifications, ensuring a robust environment for complex business operations. Developed in Quebec by Technologies Nexios TF inc., LLMUI is built for organizations requiring secure, local, and sovereign control over their language model deployments. It caters to engineering teams, data scientists, and businesses seeking advanced model orchestration, fine-tuning, and collaborative workspaces without relying on external cloud services.