← Back to community benchmarks

gemma-4-31B-it

M5 Max (40c) · 128 GB · 8bit · 2026-05-15
Performance
32k
tokens
355.4
PP tok/s
8.0
TG tok/s
92191
TTFT (ms)
39.1
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 346.7 12.3 31.8 GB view
4k 418.1 12.1 34.5 GB view
16k 423.7 10.8 35.8 GB view
32k 355.4 8.0 39.1 GB current
64k 266.1 6.5 45.1 GB view
128k 198.5 5.3 57.1 GB view
Batching Results
Batch Size TG tok/s Speedup
12.3 1.00×
16.3 1.33×
22.5 1.83×