← Back to community benchmarks
Qwen3.6-27B4bit-MTP
Performance
8k
tokens
375.5
PP tok/s
23.9
TG tok/s
21819
TTFT (ms)
18.5
Peak mem (GB)
Hardware
Chip
M5 Pro (16c)
Memory
48 GB
GPU Cores
16
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.3.2
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 364.0 | 22.8 | 16.1 GB | view |
| 4k | 391.8 | 24.4 | 17.5 GB | view |
| 8k | 375.5 | 23.9 | 18.5 GB | current |
| 16k | 356.7 | 22.2 | 20.0 GB | view |
| 32k | 334.8 | 1.7 | 23.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 22.8 | 1.00× |
| 2× | 18.1 | 0.79× |
| 4× | 7.6 | 0.33× |