← Back to community benchmarks
Qwen3.5-27B
Performance
16k
tokens
97.4
PP tok/s
13.1
TG tok/s
168267
TTFT (ms)
19.1
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
64 GB
GPU Cores
16
Software
oMLX
v0.3.6
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 100.4 | 15.3 | 15.8 GB | view |
| 4k | 101.4 | 14.9 | 17.3 GB | view |
| 16k | 97.4 | 13.1 | 19.1 GB | current |
| 64k | 80.8 | 8.9 | 26.7 GB | view |
| 128k | 65.5 | 4.9 | 37.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 15.3 | 1.00× |
| 2× | 25.1 | 1.64× |
| 4× | 26.2 | 1.71× |