← Back to community benchmarks

gpt-oss-120b-MXFP4-Q8

M5 Max (40c) · 128 GB · 4bit · 2026-05-05
Performance
4k
tokens
1,624
PP tok/s
82.6
TG tok/s
2523
TTFT (ms)
59.6
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 992.8 87.2 59.5 GB view
4k 1,624 82.6 59.6 GB current
Batching Results
Batch Size TG tok/s Speedup
87.2 1.00×
120.7 1.38×
153.5 1.76×