← Back to community benchmarks
Qwen3-VL-4B-Instruct
Performance
8k
tokens
283.5
PP tok/s
23.8
TG tok/s
28901
TTFT (ms)
4.5
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
24 GB
GPU Cores
10
Software
oMLX
v0.2.7
macOS
macOS 15.7.4
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 357.2 | 36.1 | 3.6 GB | view |
| 4k | 326.9 | 28.8 | 4.0 GB | view |
| 8k | 283.5 | 23.8 | 4.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 36.1 | 1.00× |
| 2× | 64.2 | 1.78× |
| 4× | 72.1 | 2.00× |
| 8× | 72.1 | 2.00× |