← Back to community benchmarks
Qwen3.6-27B-oQ6-mtp
Performance
32k
tokens
617.4
PP tok/s
19.9
TG tok/s
53070
TTFT (ms)
29.6
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 743.5 | 23.8 | 22.6 GB | view |
| 4k | 825.1 | 23.5 | 24.0 GB | view |
| 8k | 783.9 | 22.7 | 25.1 GB | view |
| 16k | 673.2 | 21.1 | 26.6 GB | view |
| 32k | 617.4 | 19.9 | 29.6 GB | current |
| 64k | 530.7 | 17.9 | 35.6 GB | view |
| 128k | 400.1 | 15.1 | 48.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 23.8 | 1.00× |
| 2× | 36.9 | 1.55× |
| 4× | 42.7 | 1.79× |
| 8× | 84.5 | 3.55× |