← Back to community benchmarks
Qwen3.5-9B
Performance
64k
tokens
538.3
PP tok/s
39.8
TG tok/s
121744
TTFT (ms)
10.8
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.2.9
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 631.7 | 92.0 | 6.7 GB | view |
| 4k | 638.6 | 88.1 | 7.0 GB | view |
| 8k | 633.7 | 83.3 | 7.1 GB | view |
| 16k | 620.0 | 73.8 | 7.6 GB | view |
| 32k | 591.5 | 59.5 | 8.7 GB | view |
| 64k | 538.3 | 39.8 | 10.8 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 92.0 | 1.00× |
| 2× | 118.4 | 1.29× |
| 4× | 134.7 | 1.46× |
| 8× | 147.2 | 1.60× |