← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Max (40c) · 128 GB · 4bit · 2026-05-10
Performance
64k
tokens
2,563
PP tok/s
59.0
TG tok/s
25573
TTFT (ms)
17.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,174 116.9 14.3 GB view
4k 3,363 111.6 15.0 GB view
8k 3,705 104.7 15.1 GB view
16k 3,680 94.9 15.6 GB view
32k 3,300 79.4 16.0 GB view
64k 2,563 59.0 17.7 GB current
Batching Results
Batch Size TG tok/s Speedup
116.9 1.00×
170.9 1.46×
235.9 2.02×