← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
32k
tokens
146.2
PP tok/s
22.7
TG tok/s
224068
TTFT (ms)
23.7
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.3
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 238.9 | 39.2 | 16.7 GB | view |
| 4k | 237.8 | 38.6 | 18.2 GB | view |
| 8k | 216.9 | 36.3 | 19.2 GB | view |
| 16k | 206.9 | 28.7 | 20.7 GB | view |
| 32k | 146.2 | 22.7 | 23.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 39.2 | 1.00× |
| 2× | 31.9 | 0.81× |
| 4× | 38.2 | 0.97× |