← Back to community benchmarks

GLM-5

M3 Ultra (80c) · 512 GB · 4bit · 2026-03-13
Performance
1k
tokens
187.1
PP tok/s
16.7
TG tok/s
5473
TTFT (ms)
391.8
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.11a1
macOS macOS 15.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 187.1 16.7 391.8 GB current
4k 186.4 15.3 395.1 GB view