← Back to community benchmarks

gemma-4-26b-a4b-it-UD

M5 Max (40c) · 128 GB · 4bit · 2026-05-11
Performance
16k
tokens
3,548
PP tok/s
87.1
TG tok/s
4618
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,248 104.4 15.2 GB view
4k 3,387 100.0 15.7 GB view
8k 3,602 95.4 16.2 GB view
16k 3,548 87.1 16.6 GB current
32k 3,202 76.7 17.4 GB view
64k 2,542 60.6 19.4 GB view
128k 1,758 42.9 23.5 GB view
Batching Results
Batch Size TG tok/s Speedup
104.4 1.00×
151.7 1.45×
214.7 2.06×
435.1 4.17×