Lumabri

Lumabri is a distributed private inference service that runs large AI models across local computers and servers.

Lumabri runs a private inference service across local computers and servers. It allows users to choose models, select participating computers, and approve RAM allocation to run large AI models that exceed the capacity of a single machine. The system distributes contiguous layer ranges rather than duplicating model weights across every worker's RAM. The controller, terminal interface, and inference integration are written in C, while the browser interface uses bundled HTML, CSS, and JS. Users can chat from the terminal interface or via the authenticated local web API and its bundled web page.

Lumabri at a glance

Best for: Software engineer
Runs on: Web, CLI
Pricing: Free
Released: August 2026
Category: AI coding tool
Website: github.com/justvugg/lumabri

Key features

Run supported models across multiple computers
Keep several approved models resident within explicit memory budgets
View computers, allocations, and measured workloads in a local workspace
Chat from the terminal interface or bundled web page
Configure maximum RAM and resource sharing per computer
Measure inference speed with short calibration tests

Use it for

Running large AI models that exceed single-machine memory limits
Sharing memory across local computers for private AI inference
Chatting with large language models from the terminal

Alternatives to Lumabri

Strata — Runs large AI models locally on consumer gaming PCs without server reliance.
LMDeploy — Accelerates inference with high-performance engines and quantization tools.
ODS — Automates complex local AI infrastructure setups with one command.
New Relic — Unified visibility across applications, infrastructure, and AI models.
WebLLM — Runs high-performance language models directly inside web browsers.
Mesh LLM — Run large local models across multiple machines through one OpenAI-compatible endpoint.
llm-d — Deploys production-grade LLM serving on existing hardware clusters.
Xinference — Supports a wide range of model types and hardware, including multi-node setups.
AirLLM — Runs massive 70B+ parameter models on small 4GB consumer GPUs.
Fireworks AI — Run optimized open models at lower cost and higher throughput than closed alternatives.
Foundry Local — Runs AI models locally on device with complete privacy and no Azure subscription.
fal — Access 1,000+ generative models and high-performance serverless GPUs.

All Lumabri alternatives