RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?API pricesVRAM calculatorNew

Mac vs PC for running LLMs locally

Macs win on memory; PC GPUs win on raw speed. A Mac's GPU can use most of its unified memory (about 67% under 64 GB, 75% from 64 GB by default), so a Mac M3 Ultra 512GB offers about 384 GB for a model. The biggest consumer card we track, the RTX 5090, has 32 GB but 1,792 GB/s of bandwidth, which sets how fast tokens come out.

252models at 4-bit+ on the M3 Ultra 512GB
200models at 4-bit+ on the RTX 5090
1792 GB/sRTX 5090 bandwidth
819 GB/sM3 Ultra 512GB bandwidth

Popular models: cheapest PC GPU vs smallest Mac

Fully in memory at 4-bit or better with 8K context. Speed is a bandwidth ceiling (real speed is typically 50–80% of it). When no single consumer GPU holds the model, the PC column shows the llama.cpp offload plan on the largest card instead.

ModelParams4-bit · 8KCheapest PC GPUPC tok/sSmallest MacMac tok/s
Qwen3 8B8.2B6.5 GB Arc B580 ($249)52 Mac M4 16GB18
Qwen3 14B15B10.3 GB Arc A770 16GB ($349)46 Mac M4 Pro 24GB23
gpt-oss 20B21B MoE11.8 GB Arc A770 16GB ($349)231 Mac M4 Pro 24GB99
Qwen3 Coder 30B A3B Instruct31B MoE18.7 GB RX 7900 XTX ($999)527 Mac M1 Pro 32GB110
Qwen3 32B33B21.1 GB RTX 5090 ($1999)77 Mac M3 Pro 36GB7.6
Llama 3.3 70B Instruct71B43.3 GB RTX 5090: Offload Q4_K_M, 32 GB RAM · ~1.6–3.2 tok/s— Mac M2 Max 96GB6.9
gpt-oss 120B117B MoE59.5 GB RTX 5090: Offload Q4_K_M, 64 GB RAM · ~22–44 tok/s— Mac M2 Max 96GB124
DeepSeek R1684B MoE377.8 GB no single card— none—

When a Mac makes sense

You want models bigger than a single consumer GPU can hold (70B dense, gpt-oss-120b, large MoE models) without building a multi-GPU rig, and you can accept slower prompt processing. Unified memory means no CPU-offload penalty: the whole model sits in one pool. See the Mac guide for every machine.

When a PC GPU makes sense

You mostly run models up to ~30B, want the fastest replies per dollar, or want to fine-tune. A 16–24 GB card covers most popular models at 4-bit, and mixture-of-experts models such as gpt-oss-120b can still run with llama.cpp offload if you add system RAM. See best GPU by budget and models by VRAM.

Specific models: gpt-oss-120bLlama 70BQwen3DeepSeek R1

Mac memory limits are macOS defaults (raise with iogpu.wired_limit_mb). Mac prices vary by configuration, so we compare by memory; PC prices are launch MSRPs. Methodology · Report an error.