← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Max (40c) · 128 GB · 8bit · 2026-05-23
Performance
16k
tokens
1,362
PP tok/s
68.6
TG tok/s
12031
TTFT (ms)
27.1
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,195 77.7 25.8 GB view
4k 1,406 75.1 26.3 GB view
8k 1,418 72.9 26.7 GB view
16k 1,362 68.6 27.1 GB current
32k 1,235 61.3 28.0 GB view
64k 1,025 50.7 30.0 GB view
Batching Results
Batch Size TG tok/s Speedup
77.7 1.00×
115.1 1.48×
158.3 2.04×
203.1 2.61×