← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Max (40c) · 128 GB · 8bit · 2026-06-04
Performance
16k
tokens
1,287
PP tok/s
59.6
TG tok/s
12734
TTFT (ms)
27.1
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.4.1
macOS macOS 26.5.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,118 70.4 25.8 GB view
4k 1,303 67.5 26.4 GB view
8k 1,350 66.6 26.6 GB view
16k 1,287 59.6 27.1 GB current
32k 1,171 52.7 28.0 GB view
64k 953.4 43.7 30.0 GB view
Batching Results
Batch Size TG tok/s Speedup
70.4 1.00×
96.0 1.36×
139.4 1.98×
185.6 2.64×