← Back to community benchmarks

gpt-oss-20b

M5 Pro (20c) · 64 GB · 8bit · 2026-03-28
Performance
16k
tokens
1,793
PP tok/s
45.2
TG tok/s
9137
TTFT (ms)
21.5
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.24
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,541 57.6 21.1 GB view
4k 1,905 55.2 21.2 GB view
8k 1,911 50.2 21.3 GB view
16k 1,793 45.2 21.5 GB current
32k 1,539 36.7 21.9 GB view
64k 1,128 26.0 22.6 GB view
Batching Results
Batch Size TG tok/s Speedup
57.6 1.00×
71.2 1.24×
79.3 1.38×