← Back to community benchmarks
Qwen3.5-9B
Performance
32k
tokens
636.5
PP tok/s
50.9
TG tok/s
51480
TTFT (ms)
8.7
Peak mem (GB)
Hardware
Chip
M4 Max (32c)
Memory
36 GB
GPU Cores
32
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 710.0 | 68.5 | 6.7 GB | view |
| 4k | 714.5 | 66.1 | 7.0 GB | view |
| 8k | 704.0 | 63.6 | 7.1 GB | view |
| 16k | 681.0 | 59.1 | 7.6 GB | view |
| 32k | 636.5 | 50.9 | 8.7 GB | current |
| 64k | 561.0 | 38.3 | 10.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 68.5 | 1.00× |
| 2× | 131.5 | 1.92× |
| 4× | 157.6 | 2.30× |
| 8× | 167.6 | 2.45× |