← Back to community benchmarks

MiniMax-M2.5

M3 Ultra (80c) · 256 GB · 3bit · 2026-03-14

Performance

1k

tokens

613.4

PP tok/s

50.5

TG tok/s

1669

TTFT (ms)

94.1

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 256 GB

GPU Cores 80

Software

oMLX v0.2.10

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	613.4	50.5	94.1 GB	current
4k	730.8	43.2	95.1 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	50.5	1.00×
2×	75.3	1.49×
4×	105.0	2.08×