← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · bf16 · 2026-04-13
Performance
16k
tokens
1,309
PP tok/s
25.6
TG tok/s
12517
TTFT (ms)
48.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.5.dev1
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,109 31.6 47.7 GB view
4k 1,258 29.7 48.2 GB view
8k 1,333 28.8 48.3 GB view
16k 1,309 25.6 48.7 GB current
32k 1,199 20.4 49.5 GB view
64k 968.8 14.6 51.2 GB view
Batching Results
Batch Size TG tok/s Speedup
31.6 1.00×
31.4 0.99×
51.7 1.64×
69.2 2.19×