← Back to community benchmarks
Qwen3.5-4B
Performance
1k
tokens
180.0
PP tok/s
1.8
TG tok/s
5689
TTFT (ms)
11.7
Peak mem (GB)
Hardware
Chip
M3 (10c)
Memory
24 GB
GPU Cores
10
Software
oMLX
v0.2.7
macOS
macOS 26.3
Context
1,024
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 1.8 | 1.00× |
| 2× | 2.8 | 1.56× |
| 4× | 3.9 | 2.17× |
| 8× | 9.2 | 5.11× |