can-i-run-this-llm

Hardware comparison

GPU & Mac comparisons for local LLMs

Pick any two machines from the buy list and see which runs more of the tracked open-source models — and generates tokens faster. Or start from the build advisor to find the cheapest machine for a specific model.

MacBook Air/Pro M4 (16GB)

MacBook Air M4 (24GB)

MacBook Pro M4 Pro (24GB)

MacBook Pro M4 Pro (48GB)

MacBook Pro M4 Max (48GB)

MacBook Pro M4 Max (128GB)

MacBook Air/Pro M5 (16GB)

MacBook Air/Pro M5 (24GB)

MacBook Air/Pro M5 (32GB)

MacBook Pro M5 Pro (24GB)

MacBook Pro M5 Pro (64GB)

MacBook Pro M5 Max (36GB)

MacBook Pro M5 Max (128GB)

Mac Studio M3 Ultra (96GB)

AMD Ryzen AI Max+ 395 “Strix Halo” (64GB)

AMD Ryzen AI Max+ 395 “Strix Halo” (128GB)

NVIDIA DGX Spark (GB10, 128GB)

NVIDIA RTX 3060 (12GB)

NVIDIA RTX 3090 (24GB)

NVIDIA RTX 4060 Ti (16GB)

NVIDIA RTX 4070 (12GB)

NVIDIA RTX 4080 (16GB)

NVIDIA RTX 4090 (24GB)

NVIDIA RTX 5060 Ti (16GB)

NVIDIA RTX 5070 (12GB)

NVIDIA RTX 5070 Ti (16GB)

NVIDIA RTX 5080 (16GB)

NVIDIA RTX 5090 (32GB)

NVIDIA RTX 6000 Ada (48GB)

NVIDIA RTX Pro 6000 Blackwell (96GB)

AMD RX 7900 XTX (24GB)

AMD RX 9070 (16GB)

AMD RX 9070 XT (16GB)