Syntax

Hardware support

What hardware Syntax runs on, and which capabilities each tier unlocks.

Syntax detects your hardware on first launch and chooses the right serving stack for every model you deploy. This page summarizes what's supported and what each tier unlocks.

Per-machine support matrix

HardwareOSLLM servingMultimodal servingCPU fallback
NVIDIA GPU (modern data-center, e.g., H100/H200/L40 class)Linux / Windows✓ — full coverage✓ — including image, video, audion/a
NVIDIA GPU (modern consumer, e.g., RTX 40 / 50 series)Linux / Windows✓ — most models✓ — many multimodaln/a
NVIDIA GPU (older consumer, e.g., RTX 30 / 20 series)Linux / Windows✓ — many modelspartialavailable
Apple Silicon (M1 / M2 / M3 / M4 + Pro / Max / Ultra)macOS✓ — extensive✓ — many multimodaln/a
AMD ROCm (RDNA 3 / CDNA 3 generation)Linux✓ — most modelspartialavailable
CPU only (modern x86_64 / ARM64)any✓ — smaller modelslimitedprimary

Memory guidance

For local LLM serving, plan disk and memory roughly as follows:

Model sizeDisk needed for weightsRAM (CPU)VRAM (GPU)
≤ 8B parameters~10–20 GB16 GB+8–16 GB
8–32B parameters~30–80 GB32 GB+24–48 GB
32–70B parameters~80–200 GB64 GB+48–96 GB
≥ 70B parameters~200 GB+128 GB+96 GB+ or multi-GPU

These are guidelines; actual requirements depend on the model, the quantization (when applicable), and the engine choice.

Optional dependencies

DependencyWhen it's needed
DockerOptional. Recommended on Linux when you're running engines that ship as containers. The desktop app guides you through enabling Docker if it's not present.
NVIDIA driverRequired on NVIDIA hardware. Syntax expects a recent driver; syntax doctor will warn if the version is too old.
ROCm runtimeRequired on AMD hardware. Syntax detects ROCm at first launch and falls back to CPU if it's missing.

Multi-GPU

Multi-GPU is supported on Linux for both NVIDIA and AMD where the underlying engine and the chosen model support tensor- or pipeline- parallel serving. Syntax's autotuner sets the parallelism strategy based on the model and the available GPUs without you having to pick.

Multi-host

Multi-host deployments are supported via the Remote self-hosted and Managed remote targets. For local multi-host workflows, treat each host as a remote target and deploy the party across them.

Where to go next