← Back to community benchmarks
Qwen3.6-27B-MTPLX-Optimized
Performance
16k
tokens
93.6
PP tok/s
11.5
TG tok/s
174971
TTFT (ms)
23.6
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
96 GB
GPU Cores
38
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 105.8 | 13.1 | 19.6 GB | view |
| 4k | 97.2 | 12.4 | 21.1 GB | view |
| 8k | 95.4 | 11.9 | 22.1 GB | view |
| 16k | 93.6 | 11.5 | 23.6 GB | current |
| 32k | 89.8 | 10.3 | 26.6 GB | view |
| 64k | 82.7 | 9.9 | 32.6 GB | view |
| 128k | 69.8 | 7.3 | 45.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 13.1 | 1.00× |
| 2× | 16.8 | 1.28× |
| 4× | 22.5 | 1.72× |
| 8× | 20.8 | 1.59× |