← Back to community benchmarks

gemma-4-26B-A4B-it-TurboQuant

M1 Max (32c) · 64 GB · 8bit · 2026-04-28
Performance
32k
tokens
370.7
PP tok/s
18.1
TG tok/s
88404
TTFT (ms)
27.5
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 64 GB
GPU Cores 32
Software
oMLX v0.3.8rc1
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 412.6 43.7 25.8 GB view
4k 436.6 41.6 26.2 GB view
8k 426.0 34.4 26.3 GB view
16k 399.6 26.6 26.7 GB view
32k 370.7 18.1 27.5 GB current
64k 318.7 9.1 29.2 GB view
Batching Results
Batch Size TG tok/s Speedup
43.7 1.00×
59.2 1.35×
67.0 1.53×
60.2 1.38×