← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Pro (16c) · 48 GB · 8bit · 2026-04-18
Performance
8k
tokens
1,665
PP tok/s
44.7
TG tok/s
4921
TTFT (ms)
26.3
Peak mem (GB)
Hardware
Chip M5 Pro (16c)
Memory 48 GB
GPU Cores 16
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,156 51.2 25.7 GB view
4k 1,583 49.3 26.2 GB view
8k 1,665 44.7 26.3 GB current
Batching Results
Batch Size TG tok/s Speedup
51.2 1.00×
73.3 1.43×
91.2 1.78×