← Back to community benchmarks
Qwen3___5-4B
Performance
8k
tokens
394.1
PP tok/s
34.3
TG tok/s
20789
TTFT (ms)
4.6
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 406.7 | 38.0 | 3.9 GB | view |
| 4k | 404.6 | 36.6 | 4.2 GB | view |
| 8k | 394.1 | 34.3 | 4.6 GB | current |
| 16k | 369.6 | 29.8 | 4.8 GB | view |
| 32k | 327.8 | 22.6 | 5.9 GB | view |
| 64k | 262.3 | 13.8 | 8.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 38.0 | 1.00× |
| 2× | 71.9 | 1.89× |
| 4× | 93.7 | 2.47× |