← Back to community benchmarks
Qwen3.6-27B-MTPLX-Optimized-Quality
Performance
8k
tokens
93.8
PP tok/s
9.4
TG tok/s
87354
TTFT (ms)
30.8
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
96 GB
GPU Cores
38
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 105.4 | 10.4 | 28.3 GB | view |
| 4k | 95.5 | 9.6 | 29.8 GB | view |
| 8k | 93.8 | 9.4 | 30.8 GB | current |
| 16k | 93.7 | 9.3 | 32.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 10.4 | 1.00× |
| 2× | 14.3 | 1.38× |
| 4× | 18.5 | 1.78× |