← Back to community benchmarks

Qwen2.5-32B-Instruct

M5 Max (32c) · 36 GB · 4bit · 2026-05-20
Performance
4k
tokens
573.3
PP tok/s
20.1
TG tok/s
7144
TTFT (ms)
19.3
Peak mem (GB)
Hardware
Chip M5 Max (32c)
Memory 36 GB
GPU Cores 32
Software
oMLX v0.3.9rc1
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 587.6 21.4 17.9 GB view
4k 573.3 20.1 19.3 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 21.4 1.00×
2× 37.0 1.73×
4× 45.9 2.14×