← Back to community benchmarks
Qwen3.5-27B
Performance
32k
tokens
644.6
PP tok/s
25.2
TG tok/s
50835
TTFT (ms)
20.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.9
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 640.2 | 32.6 | 16.6 GB | view |
| 4k | 695.7 | 31.9 | 17.0 GB | view |
| 8k | 717.8 | 30.6 | 17.4 GB | view |
| 16k | 704.0 | 28.4 | 18.3 GB | view |
| 32k | 644.6 | 25.2 | 20.1 GB | current |
| 64k | 528.0 | 19.9 | 23.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 32.6 | 1.00× |
| 2× | 43.3 | 1.33× |
| 4× | 61.3 | 1.88× |
| 8× | 66.5 | 2.04× |