← Back to community benchmarks
Qwen3.5-VL-9B-4bitCRACK
Performance
8k
tokens
222.0
PP tok/s
19.3
TG tok/s
36905
TTFT (ms)
7.0
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.9
macOS
macOS 26.2
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 225.2 | 21.9 | 6.5 GB | view |
| 4k | 226.1 | 21.2 | 6.8 GB | view |
| 8k | 222.0 | 19.3 | 7.0 GB | current |
| 16k | 202.9 | 16.5 | 7.5 GB | view |
| 32k | 187.5 | 12.4 | 8.5 GB | view |
| 64k | 163.8 | 8.6 | 10.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.9 | 1.00× |
| 2× | 42.3 | 1.93× |
| 4× | 50.1 | 2.29× |
| 8× | 50.4 | 2.30× |