← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Max (40c) · 64 GB · 8bit · 2026-05-13
Performance
64k
tokens
928.6
PP tok/s
50.1
TG tok/s
70573
TTFT (ms)
30.0
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 64 GB
GPU Cores 40
Software
oMLX v0.3.9.dev2
macOS macOS 26.5
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,168 78.2 25.8 GB view
4k 1,391 75.9 26.3 GB view
8k 1,370 72.7 26.8 GB view
16k 1,308 68.5 27.1 GB view
32k 1,070 60.5 28.0 GB view
64k 928.6 50.1 30.0 GB current
128k 630.6 36.3 34.2 GB view
Batching Results
Batch Size TG tok/s Speedup
78.2 1.00×
117.4 1.50×
160.0 2.05×