← Back to community benchmarks

Kimi-K2.6DQ3_K_M-q8

M3 Ultra (80c) · 512 GB · 4bit · 2026-04-22
Performance
4k
tokens
246.5
PP tok/s
21.0
TG tok/s
16620
TTFT (ms)
441.2
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.3.6
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 222.5 21.6 439.6 GB view
4k 246.5 21.0 441.2 GB current