← Back to community benchmarks

local--Agents-A1

M5 Max (40c) · 128 GB · 4bit · 2026-07-01
Performance
128k
tokens
540.2
PP tok/s
50.3
TG tok/s
242645
TTFT (ms)
22.4
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.4.5.dev1
macOS macOS 26.5.1
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 3,766 117.4 19.3 GB view
4k 2,593 119.5 20.0 GB view
8k 2,626 115.7 20.4 GB view
16k 2,008 105.8 20.3 GB view
32k 1,454 95.6 20.6 GB view
64k 967.7 75.4 21.2 GB view
128k 540.2 50.3 22.4 GB current
Batching Results
Batch Size TG tok/s Speedup
117.4 1.00×
190.9 1.63×
239.9 2.04×
289.0 2.46×