← Back to community benchmarks

gemma-4-31b-it

M3 Max (40c) · 48 GB · 4bit · 2026-04-12
Performance
4k
tokens
146.1
PP tok/s
15.0
TG tok/s
28027
TTFT (ms)
19.3
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.3.5.dev1
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 173.4 18.3 17.6 GB view
4k 146.1 15.0 19.3 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 18.3 1.00×
2× 25.7 1.40×
4× 27.0 1.48×