← Back to community benchmarks

Qwen2.5-VL-32B-Instruct

M3 Ultra (60c) · 96 GB · 8bit · 2026-05-11
Performance
1k
tokens
255.1
PP tok/s
18.3
TG tok/s
4014
TTFT (ms)
33.1
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 96 GB
GPU Cores 60
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 255.1 18.3 33.1 GB current
4k 261.9 17.6 34.6 GB view
Batching Results
Batch Size TG tok/s Speedup
1× 18.3 1.00×
2× 29.3 1.60×
4× 37.0 2.02×