← Back to community benchmarks

GLM-5

M3 Ultra (80c) · 512 GB · 4bit · 2026-03-08
Performance
16k
tokens
117.3
PP tok/s
12.0
TG tok/s
139660
TTFT (ms)
402.7
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.6
macOS macOS 15.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 186.0 16.6 391.8 GB view
4k 180.1 13.8 394.1 GB view
8k 154.2 13.1 396.7 GB view
16k 117.3 12.0 402.7 GB current
32k 77.3 10.4 415.4 GB view
Batching Results
Batch Size TG tok/s Speedup
16.6 1.00×
23.5 1.42×
30.3 1.83×
43.5 2.62×