← Back to community benchmarks

gemma-4-31B-it-PARO

M3 Ultra (60c) · 256 GB · 4bit · 2026-05-23
Performance
128k
tokens
157.3
PP tok/s
11.1
TG tok/s
833298
TTFT (ms)
42.9
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.9
macOS macOS 26.5
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 267.0 30.4 17.0 GB view
4k 267.6 29.4 19.7 GB view
8k 267.3 28.0 20.5 GB view
16k 256.4 25.7 21.7 GB view
32k 236.5 21.0 24.9 GB view
64k 204.0 15.1 30.9 GB view
128k 157.3 11.1 42.9 GB current
Batching Results
Batch Size TG tok/s Speedup
30.4 1.00×
44.5 1.46×
52.4 1.72×