← Back to community benchmarks
Qwen3.6-27B-oQ6-mtp
Performance
32k
tokens
575.7
PP tok/s
22.0
TG tok/s
56916
TTFT (ms)
29.9
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9rc1
macOS
macOS 26.5
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 739.5 | 35.1 | 23.0 GB | view |
| 4k | 721.4 | 32.2 | 24.4 GB | view |
| 8k | 670.8 | 29.0 | 25.4 GB | view |
| 16k | 641.4 | 26.1 | 26.9 GB | view |
| 32k | 575.7 | 22.0 | 29.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 35.1 | 1.00× |
| 2× | 26.9 | 0.77× |
| 4× | 30.7 | 0.87× |