← Back to community benchmarks

GLM-5

M3 Ultra (80c) · 512 GB · 4bit · 2026-03-13
Performance
4k
tokens
181.1
PP tok/s
14.0
TG tok/s
22612
TTFT (ms)
394.1
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.11a1
macOS macOS 15.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 187.4 16.8 391.8 GB view
4k 181.1 14.0 394.1 GB current