← Back to community benchmarks

gemma-4-26b-a4b-it

M3 Ultra (80c) · 512 GB · 4bit · 2026-04-17
Performance
16k
tokens
2,088
PP tok/s
61.2
TG tok/s
7848
TTFT (ms)
15.2
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.3.6
macOS macOS 15.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,479 95.2 14.2 GB view
4k 2,023 86.5 14.7 GB view
8k 2,110 75.2 14.8 GB view
16k 2,088 61.2 15.2 GB current
32k 1,881 43.2 16.0 GB view
64k 1,467 25.2 17.7 GB view
128k 1,022 14.8 21.2 GB view
195k 751.4 10.0 24.8 GB view
Batching Results
Batch Size TG tok/s Speedup
95.2 1.00×
139.4 1.46×
222.0 2.33×
302.5 3.18×