← Back to community benchmarks

gpt-oss-20b-MXFP4-Q8

M3 Max (40c) · 48 GB · 4bit · 2026-03-11
Performance
4k
tokens
932.8
PP tok/s
82.7
TG tok/s
4391
TTFT (ms)
11.5
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.2.8
macOS macOS 26.3
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 849.9 89.7 11.4 GB view
4k 932.8 82.7 11.5 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 89.7 1.00×
2× 124.6 1.39×
4× 157.9 1.76×