← Back to community benchmarks

gemma-4-26b-a4b-it

M1 Ultra (48c) · 64 GB · 4bit · 2026-04-16
Performance
32k
tokens
651.2
PP tok/s
24.0
TG tok/s
50323
TTFT (ms)
16.0
Peak mem (GB)
Hardware
Chip M1 Ultra (48c)
Memory 64 GB
GPU Cores 48
Software
oMLX v0.3.5
macOS macOS 26.3.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 587.2 61.4 14.2 GB view
4k 680.7 57.2 14.7 GB view
8k 698.9 50.3 14.8 GB view
16k 693.5 36.0 15.2 GB view
32k 651.2 24.0 16.0 GB current
Batching Results
Batch Size TG tok/s Speedup
61.4 1.00×
87.9 1.43×
114.4 1.86×