← Back to community benchmarks
Qwen3.6-27B
Performance
8k
tokens
831.9
PP tok/s
15.3
TG tok/s
9848
TTFT (ms)
35.6
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.4
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 739.7 | 15.7 | 33.0 GB | view |
| 4k | 885.8 | 15.6 | 34.6 GB | view |
| 8k | 831.9 | 15.3 | 35.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 15.7 | 1.00× |
| 2× | 20.9 | 1.33× |
| 4× | 34.1 | 2.17× |
| 8× | 49.8 | 3.17× |