← Back to community benchmarks

gpt-oss-120b

M3 Ultra (60c) · 256 GB · 8bit · 2026-03-10
Performance
4k
tokens
540.6
PP tok/s
57.6
TG tok/s
7578
TTFT (ms)
115.9
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.2.7
macOS macOS 26.3
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 484.1 64.6 115.8 GB view
4k 540.6 57.6 115.9 GB current
8k 547.2 53.5 116.1 GB view
16k 528.1 30.0 116.4 GB view
32k 484.7 18.2 117.0 GB view
64k 405.2 10.8 118.2 GB view
Batching Results
Batch Size TG tok/s Speedup
64.6 1.00×
87.6 1.36×
114.6 1.77×
140.9 2.18×