Yes — DeepSeek R2 (671B) runs at 8 tok/s on M5 Max with 128 GB RAM using Q2_K quantization via MLX. First token latency is 2.5s. A capable open-source LLM with 671B parameters.
LLMCheck measured DeepSeek R2 on M5 Max 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 | 8 tok/s |
| Time to first token | 2.5s |
| Quantization | Q2_K |
| Minimum RAM | 128 GB |
| Recommended engine | MLX |
| Parameters | 671B |
| Benchmark date | 2026-05 |
Q2_K 671B MLX M5 Max
The recommended engine for DeepSeek R2 on M5 Max is MLX. Install with pip and pull the model:
Alternatively, you can use Ollama for a simpler setup:
To run DeepSeek R2 on M5 Max you need:
See how DeepSeek R2 stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard