Yes — Qwen 3.6-27B (27B) runs at 17 tok/s on M2 Max with 32 GB RAM using Q4_K_M quantization via MLX. First token latency is 0.6s. A capable open-source LLM with 27B parameters.
Want to run Qwen 3.6-27B faster — or step up to bigger models? Find your Mac →The LLMCheck index estimates Qwen 3.6-27B on M2 Max using our published methodology: Q4_K_M quantization, memory-bandwidth scaling, and cross-referenced third-party benchmarks where available. Figures are transparent estimates — own this config? Submit a real benchmark →
| Metric | Value |
|---|---|
| Tokens per second | 17 tok/s |
| Time to first token | 0.6s |
| Quantization | Q4_K_M |
| Minimum RAM | 32 GB |
| Recommended engine | MLX |
| Parameters | 27B |
| Benchmark date | 2026-08 |
Q4_K_M 27B MLX M2 Max
The recommended engine for Qwen 3.6-27B on M2 Max is MLX. Install with pip and pull the model:
Alternatively, you can use Ollama for a simpler setup:
| Chip | Speed | First Token | Min RAM | Engine |
|---|---|---|---|---|
| M5 Max | 40 tok/s | 0.3s | 64 GB | MLX |
| M4 Max | 36 tok/s | 0.3s | 64 GB | MLX |
| M1 Ultra | 34 tok/s | 0.4s | 64 GB | MLX |
| M3 Max | 27 tok/s | 0.4s | 64 GB | Ollama |
| M5 Pro | 19 tok/s | 0.5s | 24 GB | MLX |
| M4 Pro | 18 tok/s | 0.5s | 24 GB | Ollama |
To run Qwen 3.6-27B on M2 Max you need:
Qwen 3.6-27B needs about 32 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:
As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep our benchmarks free.
See how Qwen 3.6-27B stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard