← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
1k
tokens
216.3
PP tok/s
7.9
TG tok/s
4733
TTFT (ms)
16.4
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 216.3 | 7.9 | 16.4 GB | current |
| 4k | 230.0 | 7.9 | 17.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 7.9 | 1.00× |
| 2× | 15.5 | 1.96× |
| 4× | 21.2 | 2.68× |