← Back to community benchmarks

gpt-oss-120b-MXFP4-Q8

M5 Max (40c) · 128 GB · 4bit · 2026-04-05
Performance
195k
tokens
404.2
PP tok/s
6.3
TG tok/s
494832
TTFT (ms)
92.3
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.2
macOS macOS 26.4
Context 200,000
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,023 79.7 59.5 GB view
4k 1,633 77.1 59.6 GB view
8k 1,658 70.6 59.8 GB view
16k 1,538 58.5 60.0 GB view
32k 1,213 41.4 60.6 GB view
64k 646.7 22.5 61.7 GB view
128k 521.2 14.0 64.1 GB view
195k 404.2 6.3 92.3 GB current
Batching Results
Batch Size TG tok/s Speedup
79.7 1.00×
84.7 1.06×
100.4 1.26×
111.4 1.40×