← Back to community benchmarks
Qwen3-VL-8B-Instruct
Performance
1k
tokens
224.6
PP tok/s
21.3
TG tok/s
4559
TTFT (ms)
6.0
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 224.6 | 21.3 | 6.0 GB | current |
| 4k | 217.1 | 18.9 | 6.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.3 | 1.00× |
| 2× | 41.5 | 1.95× |
| 4× | 52.9 | 2.48× |