← Back to community benchmarks
Qwen3.5-9B
Performance
195k
tokens
370.0
PP tok/s
30.1
TG tok/s
540582
TTFT (ms)
13.9
Peak mem (GB)
Hardware
Chip
M4 Max (32c)
Memory
36 GB
GPU Cores
32
Software
oMLX
v0.5.3
macOS
macOS 26.5.2
Context
200,000
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 702.6 | 66.3 | 5.8 GB | view |
| 4k | 732.1 | 64.8 | 6.3 GB | view |
| 8k | 735.6 | 63.6 | 6.5 GB | view |
| 16k | 710.7 | 60.6 | 6.8 GB | view |
| 32k | 656.8 | 55.4 | 7.3 GB | view |
| 64k | 570.8 | 47.8 | 8.6 GB | view |
| 128k | 450.9 | 37.2 | 11.2 GB | view |
| 195k | 370.0 | 30.1 | 13.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 66.3 | 1.00× |
| 2× | 123.7 | 1.87× |
| 4× | 185.0 | 2.79× |