← Back to community benchmarks
Qwen3.5-4B
Performance
4k
tokens
413.4
PP tok/s
4.6
TG tok/s
9909
TTFT (ms)
12.1
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
24 GB
GPU Cores
20
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 347.8 | 4.8 | 12.0 GB | view |
| 4k | 413.4 | 4.6 | 12.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 4.8 | 1.00× |
| 2× | 6.0 | 1.25× |
| 4× | 11.6 | 2.42× |
| 8× | 25.4 | 5.29× |