← Back to community benchmarks
Qwen3.5-9B
Performance
4k
tokens
285.7
PP tok/s
53.1
TG tok/s
14336
TTFT (ms)
7.0
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
32 GB
GPU Cores
32
Software
oMLX
v0.2.6
macOS
macOS 26.2
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 314.7 | 47.7 | 6.7 GB | view |
| 4k | 285.7 | 53.1 | 7.0 GB | current |
| 8k | 296.3 | 49.4 | 7.1 GB | view |
| 16k | 291.1 | 42.9 | 7.6 GB | view |
| 32k | 246.6 | 27.4 | 8.7 GB | view |
| 64k | 222.4 | 18.8 | 10.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 47.7 | 1.00× |
| 2× | 54.8 | 1.15× |
| 4× | 59.1 | 1.24× |
| 8× | 51.8 | 1.09× |