← Back to community benchmarks

gemma-4-26b-a4b-it-mxfp8

M3 Ultra (60c) · 256 GB · 8bit · 2026-05-13
Performance
16k
tokens
1,606
PP tok/s
69.1
TG tok/s
10204
TTFT (ms)
26.3
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.9.dev2
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,310 82.2 25.1 GB view
4k 1,633 79.1 25.7 GB view
8k 1,673 75.3 25.9 GB view
16k 1,606 69.1 26.3 GB current
32k 1,454 59.2 27.3 GB view
64k 1,189 44.6 29.3 GB view