← Back to community benchmarks

gpt-oss-120b-RotorQuant

M5 Max (40c) · 128 GB · 4bit · 2026-04-15
Performance
4k
tokens
1,643
PP tok/s
86.2
TG tok/s
2494
TTFT (ms)
58.5
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.5
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,239 92.6 58.4 GB view
4k 1,643 86.2 58.5 GB current
8k 1,450 79.5 58.6 GB view
Batching Results
Batch Size TG tok/s Speedup
92.6 1.00×
124.5 1.34×
150.8 1.63×