← Back to community benchmarks

gemma-4-26b-a4b-it-oQ8

M4 Pro (20c) · 64 GB · 8bit · 2026-04-30
Performance
4k
tokens
695.7
PP tok/s
44.3
TG tok/s
5887
TTFT (ms)
26.4
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.3
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 651.6 44.6 25.8 GB view
4k 695.7 44.3 26.4 GB current
8k 669.6 42.1 26.6 GB view
16k 611.7 39.5 27.1 GB view
Batching Results
Batch Size TG tok/s Speedup
44.6 1.00×
67.1 1.50×