← Back to community benchmarks

Ornith-1.0-35B-oQ4e

M5 Max (40c) · 128 GB · 4bit · 2026-07-13
Performance
128k
tokens
525.7
PP tok/s
44.2
TG tok/s
249331
TTFT (ms)
23.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.5.0rc1
macOS macOS 26.4
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 3,747 113.7 19.9 GB view
4k 4,478 114.6 20.8 GB view
8k 4,527 112.4 21.1 GB view
16k 2,963 102.0 21.2 GB view
32k 1,940 99.4 21.6 GB view
64k 905.7 48.4 22.5 GB view
128k 525.7 44.2 23.7 GB current
195k 338.4 33.5 25.1 GB view
Batching Results
Batch Size TG tok/s Speedup
113.7 1.00×
168.0 1.48×