← Back to community benchmarks

gemma-4-26B-A4B-it-Claude-Opus-Distill-v2-qx64-hi

M2 Max (38c) · 64 GB · 6bit · 2026-04-16
Performance
32k
tokens
558.4
PP tok/s
20.4
TG tok/s
58684
TTFT (ms)
17.1
Peak mem (GB)
Hardware
Chip M2 Max (38c)
Memory 64 GB
GPU Cores 38
Software
oMLX v0.3.5
macOS macOS 15.7.4
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 534.8 63.6 15.3 GB view
4k 598.1 57.6 15.8 GB view
8k 605.1 45.5 15.9 GB view
16k 591.2 31.8 16.3 GB view
32k 558.4 20.4 17.1 GB current
64k 490.2 0.8 21.8 GB view
128k 387.3 6.9 27.9 GB view
Batching Results
Batch Size TG tok/s Speedup
63.6 1.00×
84.8 1.33×
102.3 1.61×
120.4 1.89×