← Back to community benchmarks

gemma-4-31b-it

M1 Ultra (48c) · 128 GB · 4bit · 2026-04-05
Performance
16k
tokens
103.0
PP tok/s
4.7
TG tok/s
159035
TTFT (ms)
22.3
Peak mem (GB)
Hardware
Chip M1 Ultra (48c)
Memory 128 GB
GPU Cores 48
Software
oMLX v0.3.2
macOS macOS 15.7.2
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 107.0 15.2 18.9 GB view
4k 106.3 13.6 20.5 GB view
8k 105.4 10.8 20.9 GB view
16k 103.0 4.7 22.3 GB current
32k 98.5 4.6 24.3 GB view
Batching Results
Batch Size TG tok/s Speedup
15.2 1.00×
17.3 1.14×
19.6 1.29×