← Back to community benchmarks
Qwen3.5-9B
Performance
4k
tokens
210.9
PP tok/s
18.4
TG tok/s
19420
TTFT (ms)
6.8
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.10
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 200.6 | 17.4 | 6.5 GB | view |
| 4k | 210.9 | 18.4 | 6.8 GB | current |
| 8k | 219.2 | 17.7 | 7.0 GB | view |
| 16k | 205.7 | 16.9 | 7.5 GB | view |
| 32k | 195.1 | 13.1 | 8.5 GB | view |
| 64k | 169.9 | 9.7 | 10.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 17.4 | 1.00× |
| 2× | 36.3 | 2.09× |
| 4× | 53.9 | 3.10× |
| 8× | 54.5 | 3.13× |