← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
4k
tokens
191.8
PP tok/s
12.1
TG tok/s
21350
TTFT (ms)
18.1
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 199.3 | 12.9 | 16.6 GB | view |
| 4k | 191.8 | 12.1 | 18.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 12.9 | 1.00× |
| 2× | 25.2 | 1.95× |