← Back to community benchmarks

Qwen3-VL-30B-A3B-Instruct

M3 Ultra (60c) · 96 GB · 4bit · 2026-03-10
Performance
4k
tokens
1,881
PP tok/s
68.7
TG tok/s
2178
TTFT (ms)
18.1
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 96 GB
GPU Cores 60
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,698 78.2 17.6 GB view
4k 1,881 68.7 18.1 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 78.2 1.00×
2× 128.8 1.65×
4× 51.5 0.66×