← Back to community benchmarks
Qwen3.5-4B
Performance
16k
tokens
349.3
PP tok/s
27.4
TG tok/s
46905
TTFT (ms)
5.0
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.3.7
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 382.6 | 37.8 | 3.3 GB | view |
| 4k | 396.7 | 36.3 | 3.9 GB | view |
| 16k | 349.3 | 27.4 | 5.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 37.8 | 1.00× |
| 2× | 65.6 | 1.74× |
| 4× | 88.8 | 2.35× |