Cua releases Hoy, a macOS VMM with experimental GPU acceleration
Start here
Cua released Hoy, a closed-source macOS virtual machine monitor built on Apple's Hypervisor.framework, for agents, CI, reinforcement learning and evals that need a macOS desktop. On an M4 Pro, Cua says a Hoy VM reaches ssh in 5.3 s versus 6.6 s with Apple's Virtualization framework, and restores in 1.2 s versus 3.0 s. Hoy adds snapshot, restore and fork of running VMs. GPU acceleration for local models is experimental and off until the user turns it on. In Cua's test, llama.cpp in a Hoy VM ran at 95–97% of native prompt-processing speed on an M2 Mac. A VM cannot be snapshotted while a program inside it is using the GPU, though restore and fork still work. Hoy is free, including for commercial use. It sends crash reports and basic usage counts by default; each has its own switch, via `hoy config set crash-reports off` and `hoy config set usage-data off`.
Generated from reports · updated 7 h ago
Report timeline
Follow the reports to see the event from different sides.
- Hacker News · AICua releases Hoy, a macOS VMM on Hypervisor.framework with GPU acceleration for local models
Cua released Hoy, a closed-source macOS virtual machine monitor built on Apple's Hypervisor.framework, aimed at agents, CI, reinforcement learning and evals that need a macOS desktop. On an M4 Pro, a Hoy VM reaches ssh in 5.3 s versus 6.6 s with Apple's Virtualization framework, and restores in 1.2 s versus 3.0 s. Hoy adds snapshot, restore and fork of running VMs, and an experimental GPU mode where llama.cpp in a Hoy VM ran at 95 to 97% of native prompt-processing speed on an M2 Mac.
Heat for this event
- Current, comparable range
- 8
- Peak, comparable range
- 10November 18 01:00
- Change in 24 hours
- –
The trend compares the same participants over hours observed continuously, so its range can be narrower than the current heat count. Hover or tap the chart for hourly heat; the arrow keys move through it.