← Back to community benchmarks

gemma-4-26b-a4b-it-UD

M5 Max (40c) · 128 GB · 4bit · 2026-05-11
Performance
32k
tokens
3,206
PP tok/s
76.5
TG tok/s
10219
TTFT (ms)
17.4
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 26.5
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,193 105.2 15.2 GB view
4k 3,342 100.5 15.7 GB view
8k 3,514 92.3 16.2 GB view
16k 3,505 86.5 16.6 GB view
32k 3,206 76.5 17.4 GB current
64k 2,539 61.0 19.4 GB view
128k 1,724 42.9 23.5 GB view
Batching Results
Batch Size TG tok/s Speedup
105.2 1.00×
152.6 1.45×
213.4 2.03×
436.8 4.15×