Open-source distributed inference project. Builds "exo," a tool that turns Macs and consumer devices into a local LLM cluster using topology-aware model sharding, MLX as the inference backend, and RDMA over Thunderbolt 5 for a claimed 99% latency reduction between nodes. Supports LLaMA, DeepSeek, and Qwen via OpenAI-compatible APIs.
EXO Labs is building a free, transparent benchmark site covering every LLM, quantization level, and consumer hardware setup — so you can shop for AI compute with real data.