← Back to community benchmarks

community--Ornith-1.0-35B-5bit

M3 Max (40c) · 48 GB · 5bit · 2026-07-24

Performance

64k

tokens

789.0

PP tok/s

51.3

TG tok/s

83060

TTFT (ms)

25.2

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 48 GB

GPU Cores 40

Software

oMLX v0.5.1

macOS macOS 26.5.2

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
64k	789.0	51.3	25.2 GB	current
128k	457.5	25.4	28.2 GB	view
195k	197.5	25.5	29.9 GB	view