← Back to community benchmarks
Qwen3.5-9B-OptiQ
Performance
32k
tokens
592.1
PP tok/s
53.9
TG tok/s
55345
TTFT (ms)
10.1
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.3.4
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 552.7 | 90.8 | 6.8 GB | view |
| 4k | 622.0 | 85.8 | 7.5 GB | view |
| 8k | 627.3 | 81.2 | 7.8 GB | view |
| 16k | 619.0 | 65.5 | 8.6 GB | view |
| 32k | 592.1 | 53.9 | 10.1 GB | current |
| 64k | 533.4 | 38.5 | 13.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 90.8 | 1.00× |
| 2× | 117.7 | 1.30× |
| 4× | 139.1 | 1.53× |