← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Max (40c) · 128 GB · 4bit · 2026-04-03
Performance
16k
tokens
1,327
PP tok/s
54.5
TG tok/s
12346
TTFT (ms)
16.5
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.2
macOS macOS 15.7.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,433 71.3 15.3 GB view
4k 1,424 67.2 15.9 GB view
8k 1,398 62.4 16.1 GB view
16k 1,327 54.5 16.5 GB current
32k 1,195 44.0 17.2 GB view
64k 983.3 30.8 18.9 GB view
128k 696.9 1.0 22.5 GB view
Batching Results
Batch Size TG tok/s Speedup
71.3 1.00×
105.2 1.48×
152.4 2.14×