← Back to community benchmarks
Qwen3.5-9B
Performance
32k
tokens
296.0
PP tok/s
37.7
TG tok/s
110688
TTFT (ms)
8.7
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 316.1 | 56.2 | 6.7 GB | view |
| 4k | 319.2 | 53.6 | 7.0 GB | view |
| 8k | 316.7 | 51.1 | 7.1 GB | view |
| 16k | 309.1 | 45.8 | 7.6 GB | view |
| 32k | 296.0 | 37.7 | 8.7 GB | current |
| 64k | 268.1 | 30.6 | 10.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 56.2 | 1.00× |
| 2× | 67.0 | 1.19× |
| 4× | 72.4 | 1.29× |