← Back to community benchmarks

dolphin-2.9.1-llama-3-70b

M3 Max (40c) · 128 GB · 8bit · 2026-03-10

Performance

1k

tokens

83.6

PP tok/s

4.9

TG tok/s

12266

TTFT (ms)

70.6

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.2.6

macOS macOS 26.2

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	83.6	4.9	70.6 GB	current
4k	69.0	4.8	71.9 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	4.9	1.00×
2×	9.5	1.94×
4×	12.8	2.61×