← Back to community benchmarks
Qwen3.6-27B-UD
Performance
32k
tokens
698.9
PP tok/s
14.0
TG tok/s
46882
TTFT (ms)
31.0
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.7
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 670.6 | 21.1 | 25.2 GB | view |
| 4k | 843.6 | 20.8 | 26.6 GB | view |
| 8k | 901.2 | 20.4 | 27.3 GB | view |
| 16k | 842.3 | 16.5 | 28.5 GB | view |
| 32k | 698.9 | 14.0 | 31.0 GB | current |
| 64k | 538.6 | 12.4 | 36.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.1 | 1.00× |
| 2× | 23.0 | 1.09× |
| 4× | 37.5 | 1.78× |