← Back to community benchmarks

gpt-oss-120b-MXFP4-Q8

M3 Ultra (80c) · 512 GB · 4bit · 2026-03-11
Performance
4k
tokens
703.2
PP tok/s
67.6
TG tok/s
5825
TTFT (ms)
59.3
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 583.5 65.4 59.2 GB view
4k 703.2 67.6 59.3 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 65.4 1.00×
2× 108.3 1.66×
4× 145.2 2.22×