← Back to community benchmarks
Qwen3.5-9B
Performance
1k
tokens
336.2
PP tok/s
25.3
TG tok/s
3046
TTFT (ms)
6.7
Peak mem (GB)
Hardware
Chip
M3 Pro (18c)
Memory
18 GB
GPU Cores
18
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 336.2 | 25.3 | 6.7 GB | current |
| 4k | 347.4 | 25.0 | 7.0 GB | view |
| 8k | 342.7 | 24.3 | 7.1 GB | view |
| 16k | 330.3 | 22.6 | 7.6 GB | view |
| 32k | 305.7 | 19.1 | 8.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 25.3 | 1.00× |
| 2× | 51.6 | 2.04× |
| 4× | 81.1 | 3.21× |
| 8× | 83.2 | 3.29× |