← Back to community benchmarks
Qwen3.5-4B
Performance
4k
tokens
398.7
PP tok/s
35.8
TG tok/s
10274
TTFT (ms)
4.2
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
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 | 405.7 | 36.8 | 3.9 GB | view |
| 4k | 398.7 | 35.8 | 4.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 36.8 | 1.00× |
| 2× | 70.1 | 1.90× |
| 4× | 90.3 | 2.45× |