← Back to community benchmarks

gemma-4-26B-A4B-it-heretic-msq-2.6bit

M5 (10c) · 16 GB · 4bit · 2026-04-30
Performance
1k
tokens
368.4
PP tok/s
37.5
TG tok/s
2780
TTFT (ms)
9.1
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.3.5
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 368.4 37.5 9.1 GB current
4k 381.2 34.2 9.6 GB view
Batching Results
Batch Size TG tok/s Speedup
37.5 1.00×
54.3 1.45×
51.9 1.38×