Yes — Llama 5 405B (405B) runs at 5 tok/s on M4 Ultra with 192 GB RAM using Q2_K quantization via MLX. First token latency is 3.0s. A capable open-source LLM with 405B parameters.
LLMCheck measured Llama 5 405B on M4 Ultra using the standard methodology: Q4_K_M quantization, 256-token input, 512-token output, 3 runs averaged on a freshly-booted system.
| Metric | Value |
|---|---|
| Tokens per second | 5 tok/s |
| Time to first token | 3.0s |
| Quantization | Q2_K |
| Minimum RAM | 192 GB |
| Recommended engine | MLX |
| Parameters | 405B |
| Benchmark date | 2026-07 |
Q2_K 405B MLX M4 Ultra
The recommended engine for Llama 5 405B on M4 Ultra is MLX. Install with pip and pull the model:
Alternatively, you can use Ollama for a simpler setup:
To run Llama 5 405B on M4 Ultra you need:
Llama 5 405B needs about 192 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:
Not sure which Mac fits your budget? See the best Mac for running this →
As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep our benchmarks free.
See how Llama 5 405B stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard