← Back to community benchmarks

gemma-4-26b-a4b-it

M1 Max (32c) · 32 GB · 4bit · 2026-05-12
Performance
16k
tokens
476.8
PP tok/s
43.5
TG tok/s
34361
TTFT (ms)
15.6
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 32 GB
GPU Cores 32
Software
oMLX v0.3.8
macOS macOS 15.6.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 425.0 55.2 14.3 GB view
4k 471.2 52.0 14.9 GB view
8k 487.0 49.1 15.1 GB view
16k 476.8 43.5 15.6 GB current
32k 447.8 35.3 16.5 GB view
64k 392.5 22.6 18.5 GB view
128k 294.4 13.6 21.2 GB view
Batching Results
Batch Size TG tok/s Speedup
55.2 1.00×
72.2 1.31×
87.1 1.58×