← Back to community benchmarks
Qwen3.5-VL-4B-4bitCRACK
Performance
1k
tokens
628.8
PP tok/s
80.9
TG tok/s
1628
TTFT (ms)
4.0
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
48 GB
GPU Cores
16
Software
oMLX
v0.2.7
macOS
macOS 26.4
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 628.8 | 80.9 | 4.0 GB | current |
| 4k | 642.0 | 76.4 | 4.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 80.9 | 1.00× |
| 2× | 129.8 | 1.60× |
| 4× | 143.1 | 1.77× |
| 8× | 141.8 | 1.75× |