← Back to community benchmarks
Qwen3-VL-4B-Instruct
Performance
1k
tokens
398.1
PP tok/s
35.7
TG tok/s
2572
TTFT (ms)
3.6
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.4
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 398.1 | 35.7 | 3.6 GB | current |
| 4k | 341.7 | 31.0 | 4.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 35.7 | 1.00× |
| 2× | 70.3 | 1.97× |
| 4× | 83.6 | 2.34× |