← Back to community benchmarks

Kimi-K2.6DQ3_K_M-q8

M3 Ultra (80c) · 512 GB · 4bit · 2026-05-25

Performance

1k

tokens

222.8

PP tok/s

22.1

TG tok/s

4595

TTFT (ms)

439.6

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.10

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	222.8	22.1	439.6 GB	current
4k	248.4	21.8	441.2 GB	view
8k	232.1	21.0	443.7 GB	view
16k	202.9	18.6	448.8 GB	view
32k	157.6	16.3	458.9 GB	view