← Back to community benchmarks
Qwen3.5-4B
Performance
32k
tokens
924.4
PP tok/s
61.1
TG tok/s
35449
TTFT (ms)
6.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,115 | 128.0 | 3.4 GB | view |
| 4k | 1,307 | 118.1 | 3.5 GB | view |
| 8k | 1,304 | 100.7 | 3.9 GB | view |
| 16k | 1,168 | 82.9 | 4.7 GB | view |
| 32k | 924.4 | 61.1 | 6.3 GB | current |
| 64k | 686.3 | 39.1 | 8.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 128.0 | 1.00× |
| 2× | 190.8 | 1.49× |
| 4× | 240.9 | 1.88× |
| 8× | 169.0 | 1.32× |