← Back to community benchmarks

Qwen3-30B-A3B-Instruct-2507

M2 Max (38c) · 64 GB · 8bit · 2026-03-11
Performance
1k
tokens
547.3
PP tok/s
59.5
TG tok/s
1871
TTFT (ms)
30.8
Peak mem (GB)
Hardware
Chip M2 Max (38c)
Memory 64 GB
GPU Cores 38
Software
oMLX v0.2.7
macOS macOS 15.6.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 547.3 59.5 30.8 GB current
4k 523.7 48.5 31.3 GB view
Batching Results
Batch Size TG tok/s Speedup
59.5 1.00×
73.0 1.23×
83.3 1.40×