← Back to community benchmarks

MiniMax-M2.7-RAM-100GB

M5 Max (40c) · 128 GB · 4bit · 2026-04-12

Performance

4k

tokens

826.2

PP tok/s

28.5

TG tok/s

4958

TTFT (ms)

95.0

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.5.dev2

macOS macOS 26.5

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	639.4	48.4	94.1 GB	view
4k	826.2	28.5	95.0 GB	current
8k	741.9	25.9	96.0 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	48.4	1.00×
2×	47.3	0.98×
4×	54.5	1.13×