← Back to community benchmarks

gemma-4-26b-a4b-it

M2 Max (30c) · 64 GB · 8bit · 2026-05-15
Performance
32k
tokens
460.2
PP tok/s
36.5
TG tok/s
71196
TTFT (ms)
28.0
Peak mem (GB)
Hardware
Chip M2 Max (30c)
Memory 64 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 435.5 53.6 25.8 GB view
4k 482.7 51.2 26.4 GB view
8k 503.3 48.5 26.6 GB view
16k 490.5 44.3 27.1 GB view
32k 460.2 36.5 28.0 GB current
64k 403.3 26.9 30.0 GB view
128k 323.0 17.8 32.7 GB view