← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
8k
tokens
247.7
PP tok/s
21.1
TG tok/s
33071
TTFT (ms)
18.9
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.5
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 239.5 | 27.3 | 16.4 GB | view |
| 4k | 249.6 | 25.9 | 17.9 GB | view |
| 8k | 247.7 | 21.1 | 18.9 GB | current |
| 16k | 240.8 | 22.9 | 20.4 GB | view |
| 32k | 227.1 | 18.0 | 23.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 27.3 | 1.00× |
| 2× | 32.7 | 1.20× |
| 4× | 35.2 | 1.29× |
| 8× | 55.5 | 2.03× |