← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Max (40c) · 128 GB · 8bit · 2026-04-03
Performance
64k
tokens
973.5
PP tok/s
25.5
TG tok/s
67318
TTFT (ms)
30.4
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 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,399 56.5 26.8 GB view
4k 1,398 54.5 27.4 GB view
8k 1,371 51.1 27.6 GB view
16k 1,313 46.6 28.0 GB view
32k 1,187 38.7 28.7 GB view
64k 973.5 25.5 30.4 GB current
128k 697.2 0.6 34.0 GB view
Batching Results
Batch Size TG tok/s Speedup
56.5 1.00×
82.9 1.47×
124.3 2.20×