← Back to community benchmarks

MiniMax-M2.5

M3 Ultra (80c) · 512 GB · 8bit · 2026-03-15

Performance

8k

tokens

673.0

PP tok/s

27.0

TG tok/s

12173

TTFT (ms)

229.1

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.2.12

macOS macOS 26.3.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	571.9	34.8	227.2 GB	view
4k	710.1	31.3	228.2 GB	view
8k	673.0	27.0	229.1 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	34.8	1.00×
2×	51.3	1.47×
4×	66.8	1.92×