← Back to community benchmarks

gpt-oss-20b-MXFP4-Q8

M4 (10c) · 24 GB · 4bit · 2026-04-16
Performance
1k
tokens
375.3
PP tok/s
35.1
TG tok/s
2728
TTFT (ms)
11.7
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.5
macOS macOS 26.2
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 375.3 35.1 11.7 GB current
4k 401.5 32.8 11.8 GB view
Batching Results
Batch Size TG tok/s Speedup
35.1 1.00×
46.4 1.32×
54.9 1.56×