← Back to community benchmarks

GLM-5

M3 Ultra (80c) · 512 GB · 4bit · 2026-03-13
Performance
8k
tokens
163.0
PP tok/s
14.9
TG tok/s
50258
TTFT (ms)
398.0
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.11a1
macOS macOS 15.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
8k 163.0 14.9 398.0 GB current
16k 127.2 14.1 404.0 GB view