Hello Exo Team,
I am an AI architecture enthusiast currently building a unified, heterogeneous local inference cluster at home. I am pushing the limits of Exo by distributing LLMs across a diverse mix of architectures and operating systems using Pipeline Parallelism.
My current hardware setup for the cluster includes:
- Mac Mini (Apple Silicon - acting as the primary orchestrator)
- Intel-based Mac (x86 architecture)
- Windows PC (Intel-based processor)
- iPad Pro & multiple iPhones (Intended as iOS compute nodes)
I've been successfully experimenting with the desktop nodes, but to complete my architectural pipeline and truly test the cross-platform capabilities of Exo, I need to bring my iPad and iPhones into the unified memory pool natively.
Could you please invite me to the iOS TestFlight beta?
I would love to contribute by testing the network profiling, payload distribution, and overall latency across this highly heterogeneous (ARM + x86 / macOS + Windows + iOS) environment, and provide detailed feedback.
Thank you for this amazing project!
Hello Exo Team,
I am an AI architecture enthusiast currently building a unified, heterogeneous local inference cluster at home. I am pushing the limits of Exo by distributing LLMs across a diverse mix of architectures and operating systems using Pipeline Parallelism.
My current hardware setup for the cluster includes:
I've been successfully experimenting with the desktop nodes, but to complete my architectural pipeline and truly test the cross-platform capabilities of Exo, I need to bring my iPad and iPhones into the unified memory pool natively.
Could you please invite me to the iOS TestFlight beta?
I would love to contribute by testing the network profiling, payload distribution, and overall latency across this highly heterogeneous (ARM + x86 / macOS + Windows + iOS) environment, and provide detailed feedback.
Thank you for this amazing project!