← Back to community benchmarks

gpt-oss-20b-MXFP4-Q8

M4 (10c) · 24 GB · 4bit · 2026-04-07
Performance
32k
tokens
261.8
PP tok/s
5.3
TG tok/s
125173
TTFT (ms)
12.4
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.4
macOS macOS 26.4
Context 32,768