← Back to community benchmarks
Qwen3.5-VL-2B-4bitCRACK
Performance
4k
tokens
1,044
PP tok/s
82.3
TG tok/s
3923
TTFT (ms)
2.5
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.7
macOS
macOS 26.2
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,045 | 86.8 | 2.5 GB | view |
| 4k | 1,044 | 82.3 | 2.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 86.8 | 1.00× |
| 2× | 162.9 | 1.88× |
| 4× | 206.9 | 2.38× |
| 8× | 212.0 | 2.44× |