🖥️ Multi-GPU Optimized
Automatically detects NVIDIA GPUs and supports multi-GPU model splitting and dual-model setups (main chat + lightweight helper).
👤 Modern Dashboard
Streaming chat UI with Markdown + math, live server logs, NVIDIA SMI, and performance analytics built in.
⚙️ Model Management
Scan your model library, start/stop/switch instantly, and tune runtime settings without editing configs.
🔒 Runs Locally
No cloud required. Run on your own hardware with full privacy and LAN-first control.
🧩 CE + Pro Roadmap
Community Edition focuses on a solid local experience. Pro will add multi-server orchestration and scale-out features.
🐧 Linux & Windows Support
Built to run on real machines: Windows and Linux deployments with broad NVIDIA hardware support.
LLM Controller
Beta 4
About LLM Controller
LLM Controller is a local-first dashboard for running and managing Large Language Models on your own hardware.
Launch, switch, and monitor models with zero cloud dependencies, full privacy, and real-time insight into performance and GPU usage.
Built on llama-server, it detects NVIDIA GPUs and supports advanced multi-GPU and dual-model setups without manual configuration.
Key Features
- Model Management — scan, launch, stop, and switch models quickly.
- Live Analytics — logs, NVIDIA SMI, throughput, and latency metrics.
- Modern Chat UI — streaming output, Markdown, code, math, and titles.
- Local & Private — fully self-hosted. No cloud, no data sharing.
- Actively Developed — built to evolve with new models and features.
LLM Controller CE (Community Edition)
Local AI, done right.