← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Max (40c) · 128 GB · 4bit · 2026-05-09
Performance
32k
tokens
475.2
PP tok/s
32.3
TG tok/s
68950
TTFT (ms)
16.5
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 543.4 49.7 14.3 GB view
4k 597.9 46.4 14.9 GB view
8k 597.7 43.9 15.1 GB view
16k 551.7 38.7 15.6 GB view
32k 475.2 32.3 16.5 GB current
64k 372.7 26.7 18.5 GB view
128k 241.0 18.5 22.7 GB view
195k 174.0 11.5 27.7 GB view
Batching Results
Batch Size TG tok/s Speedup
49.7 1.00×
51.6 1.04×
68.0 1.37×
78.6 1.58×