← Back to community benchmarks

gemma-4-31b-it-mxfp8

M3 Ultra (60c) · 256 GB · 8bit · 2026-05-10
Performance
16k
tokens
239.1
PP tok/s
16.7
TG tok/s
68516
TTFT (ms)
35.5
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.9.dev1
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 245.7 18.3 30.9 GB view
4k 249.1 17.9 33.6 GB view
8k 247.3 17.5 34.1 GB view
16k 239.1 16.7 35.5 GB current
32k 223.3 15.2 38.3 GB view
64k 195.1 12.9 44.9 GB view
128k 152.1 9.1 56.2 GB view
195k 123.0 7.5 71.3 GB view
Batching Results
Batch Size TG tok/s Speedup
18.3 1.00×
24.3 1.33×
26.6 1.45×
53.2 2.91×