← Back to community benchmarks
Qwen3.5-27B-oQ4
Performance
4k
tokens
333.3
PP tok/s
12.2
TG tok/s
12289
TTFT (ms)
21.9
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
48 GB
GPU Cores
20
Software
oMLX
v0.2.20
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 324.6 | 12.5 | 18.8 GB | view |
| 4k | 333.3 | 12.2 | 21.9 GB | current |
| 8k | 324.8 | 11.7 | 23.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 12.5 | 1.00× |
| 2× | 22.9 | 1.83× |
| 4× | 33.8 | 2.70× |
| 8× | 38.7 | 3.10× |