Far Labs and Evolving Edge launch GPU rental services for AI

Startups Far Labs and Evolving Edge are developing distributed computing markets that allow users to rent out the idle GPU power from their gaming PCs to support AI workloads.

Far Labs, based in Abu Dhabi, aims to launch its Far AI platform within weeks, targeting latency levels below 100 ms. CEO Ilman Shazhaev describes the concept as an ‘Uber or Airbnb for AI inference tasks.’ Evolving Edge, located in Austin, has already opened a testing program for users. Both companies focus on small to medium-sized open-source models rather than the massive systems used by companies like OpenAI or Google.

These platforms allow AI companies to utilize existing consumer hardware for inference—the process of running a trained model to generate results—without building massive dedicated data centers. This model competes with existing networks like Salad, which reportedly operates over 60,000 consumer GPUs daily.