Computers

Which kind of computer do you have?

We sort projects into a few groups by what a computer can comfortably handle. Find yours below.

Basic laptop (8 GB)

An everyday laptop with about 8 GB of memory and no separate graphics card. It can handle the lighter projects, like document search and short transcriptions.

8 GB RAM, CPU only, integrated graphics. Our test machines: Surface laptop (i5-8250U, 8 GB).

Regular laptop or desktop (16 GB)

The most common setup today. Enough for a small private chat assistant and most everyday projects. Answers come at a relaxed pace.

16 GB RAM, CPU only. Our test machines: Desktop (i5-6600, 16 GB), Surface laptop (i7-1185G7, 16 GB).

Gaming PC (NVIDIA graphics card)

A desktop with an NVIDIA graphics card. The card does the heavy lifting, so chat and transcription run much faster and bigger models fit.

NVIDIA GPU with about 16 GB VRAM (CUDA). Our test machines: Gaming PC (RTX 4060 Ti 16 GB, Ryzen 7 1800X, 64 GB).

Newer Mac (32 GB or more)

An Apple Silicon Mac (M1 through M4) with plenty of memory. Macs share memory between the processor and graphics, which makes them very good at this.

Apple Silicon, 32 to 48 GB unified memory, Metal / MLX. Our test machines: M3 Max 36 GB, M4 Pro 48 GB.

AI workstations Coming later

Dedicated machines like NVIDIA's DGX Spark or a high-memory Mac Studio. We'll cover these once we can test one properly.

96 GB+ unified memory.

How to check what you have

• Windows: Settings > System > About shows "Installed RAM". For a graphics card, open Task Manager > Performance and look for "GPU".
• Mac: Apple menu > About This Mac shows the chip (M1, M2, M3, M4) and memory.
• Linux: run free -h for memory and nvidia-smi for an NVIDIA card.

Our test computers