← Back to community benchmarks
Qwen3.5-27B
Performance
32k
tokens
170.0
PP tok/s
11.0
TG tok/s
192790
TTFT (ms)
35.2
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.2.18
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 247.0 | 16.7 | 29.2 GB | view |
| 4k | 219.6 | 16.2 | 30.8 GB | view |
| 8k | 209.1 | 15.6 | 31.4 GB | view |
| 16k | 208.0 | 13.2 | 32.7 GB | view |
| 32k | 170.0 | 11.0 | 35.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 16.7 | 1.00× |
| 2× | 24.4 | 1.46× |
| 4× | 29.9 | 1.79× |
| 8× | 35.3 | 2.11× |