← Back to community benchmarks
Qwen3-VL-8B-Instruct
Performance
16k
tokens
159.8
PP tok/s
7.2
TG tok/s
102525
TTFT (ms)
11.9
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 217.2 | 11.3 | 9.7 GB | view |
| 4k | 198.4 | 9.9 | 10.3 GB | view |
| 8k | 183.1 | 8.9 | 10.8 GB | view |
| 16k | 159.8 | 7.2 | 11.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 11.3 | 1.00× |
| 2× | 20.9 | 1.85× |
| 4× | 33.3 | 2.95× |
| 8× | 34.6 | 3.06× |