← Back to community benchmarks

gemma-4-26b-a4b-it

M3 Ultra (60c) · 256 GB · bf16 · 2026-07-13
Performance
32k
tokens
1,354
PP tok/s
48.8
TG tok/s
24199
TTFT (ms)
53.9
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.5.1
macOS macOS 26.5.2
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,699 62.1 47.8 GB view
4k 1,630 60.4 48.1 GB view
8k 1,589 57.8 48.7 GB view
16k 1,504 55.2 50.5 GB view
32k 1,354 48.8 53.9 GB current
64k 1,111 39.4 60.8 GB view
128k 792.1 27.8 74.5 GB view
Batching Results
Batch Size TG tok/s Speedup
62.1 1.00×
56.8 0.91×
97.3 1.57×