← Back to community benchmarks

Qwen3-VL-30B-A3B-Instruct

M4 Pro (20c) · 48 GB · 4bit · 2026-03-10
Performance
8k
tokens
672.0
PP tok/s
48.0
TG tok/s
12191
TTFT (ms)
18.4
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.2.6
macOS macOS 26.3
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 838.1 66.4 17.6 GB view
4k 779.7 58.0 18.1 GB view
8k 672.0 48.0 18.4 GB current
16k 507.7 36.3 19.2 GB view
32k 338.0 22.8 20.6 GB view
64k 199.4 8.2 23.6 GB view
Batching Results
Batch Size TG tok/s Speedup
66.4 1.00×
95.0 1.43×
120.3 1.81×
156.2 2.35×