← Back to community benchmarks

gemma-4-26b-a4b-it

M3 Max (30c) · 36 GB · 4bit · 2026-05-12
User commentmlx-community/gemma-4-26b-a4b-it-4bit
Performance
32k
tokens
749.8
PP tok/s
49.8
TG tok/s
43702
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip M3 Max (30c)
Memory 36 GB
GPU Cores 30
Software
oMLX v0.3.8
macOS macOS 26.5
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 803.9 71.4 14.3 GB view
4k 930.2 68.6 15.0 GB view
8k 930.7 64.1 15.2 GB view
16k 873.5 58.5 15.6 GB view
32k 749.8 49.8 16.6 GB current
64k 597.9 36.1 17.7 GB view
128k 370.1 10.0 21.2 GB view
Batching Results
Batch Size TG tok/s Speedup
71.4 1.00×
83.3 1.17×
96.1 1.35×
190.6 2.67×