> For the complete documentation index, see [llms.txt](https://docs.nvcl.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nvcl.ai/reference/hardware.md).

# Hardware

NVCL AI supports NVIDIA H100, H200, B200, B300 and RTX PRO 6000 GPUs, and Macs with Apple Silicon (M1 or newer).

* **Two cards in one server:** one model split across both cards (TP2) or one copy per card (DP2). Which to choose: [GPU layout](/get-started/install.md#gpu-layout); results: [Two H200](/benchmarks/two-gpus.md).
* **Operating system:** Linux on x86-64 with glibc 2.35 or newer (Ubuntu 22.04, Debian 12, RHEL 10 or newer); on a Mac, macOS 13.3 or newer.
* **Mac:** Qwen3.8-27B only, in unified memory; what it needs: [Requirements](/get-started/requirements.md#mac).
* **What each model needs** (GPUs, GPU memory, host memory, disk, driver): [Requirements](/get-started/requirements.md).

Our benchmark results were measured on H200: [Benchmarks](/benchmarks/benchmarks.md).