← Back to community benchmarks

gemma-4-26B-A4B-it-mxfp4

M5 (10c) · 24 GB · 4bit · 2026-04-03
Performance
4k
tokens
986.7
PP tok/s
36.0
TG tok/s
4151
TTFT (ms)
14.7
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.1
macOS macOS 26.2
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 917.2 35.3 14.3 GB view
4k 986.7 36.0 14.7 GB current
Batching Results
Batch Size TG tok/s Speedup
35.3 1.00×
43.8 1.24×
56.9 1.61×