← Back to community benchmarks
Qwen3.5-4B
Performance
32k
tokens
322.8
PP tok/s
23.1
TG tok/s
101507
TTFT (ms)
5.9
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 406.1 | 37.8 | 3.9 GB | view |
| 8k | 392.3 | 34.0 | 4.6 GB | view |
| 16k | 368.5 | 29.6 | 4.8 GB | view |
| 32k | 322.8 | 23.1 | 5.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 37.8 | 1.00× |
| 2× | 73.5 | 1.94× |
| 4× | 91.6 | 2.42× |
| 8× | 94.2 | 2.49× |