← Back to community benchmarks
Qwen3.5-4B
Performance
32k
tokens
280.6
PP tok/s
30.5
TG tok/s
116764
TTFT (ms)
6.9
Peak mem (GB)
Hardware
Chip
M2 Pro (16c)
Memory
16 GB
GPU Cores
16
Software
oMLX
v0.2.7
macOS
macOS 26.3
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 324.3 | 59.8 | 4.0 GB | view |
| 4k | 323.2 | 54.6 | 4.2 GB | view |
| 8k | 316.8 | 49.9 | 4.6 GB | view |
| 16k | 304.0 | 41.2 | 5.3 GB | view |
| 32k | 280.6 | 30.5 | 6.9 GB | current |
| 64k | 222.3 | 16.5 | 8.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 59.8 | 1.00× |
| 2× | 67.7 | 1.13× |
| 4× | 74.9 | 1.25× |
| 8× | 64.5 | 1.08× |