← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
16k
tokens
93.6
PP tok/s
11.1
TG tok/s
174952
TTFT (ms)
20.7
Peak mem (GB)
Hardware
Chip
M3 Pro (18c)
Memory
36 GB
GPU Cores
18
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 95.9 | 13.6 | 16.7 GB | view |
| 4k | 96.4 | 13.7 | 18.2 GB | view |
| 8k | 95.4 | 12.4 | 19.2 GB | view |
| 16k | 93.6 | 11.1 | 20.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 13.6 | 1.00× |
| 2× | 16.5 | 1.21× |
| 4× | 22.3 | 1.64× |
| 8× | 23.2 | 1.71× |