← Back to community benchmarks
Qwen3.5-9B
Performance
32k
tokens
184.0
PP tok/s
15.6
TG tok/s
178106
TTFT (ms)
8.9
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 214.8 | 21.3 | 5.7 GB | view |
| 16k | 198.5 | 15.0 | 7.4 GB | view |
| 32k | 184.0 | 15.6 | 8.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.3 | 1.00× |
| 2× | 41.6 | 1.95× |
| 4× | 51.8 | 2.43× |