← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
16k
tokens
180.3
PP tok/s
18.1
TG tok/s
90879
TTFT (ms)
20.1
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9rc1
macOS
macOS 15.7.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 207.6 | 21.8 | 16.4 GB | view |
| 4k | 195.5 | 19.2 | 17.9 GB | view |
| 16k | 180.3 | 18.1 | 20.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.8 | 1.00× |
| 2× | 28.7 | 1.32× |
| 4× | 37.0 | 1.70× |
| 8× | 37.1 | 1.70× |