← Back to community benchmarks

Qwen2.5-Coder-14B-Instruct

M5 Max (40c) · 128 GB · 4bit · 2026-04-16

Performance

1k

tokens

1,556

PP tok/s

56.8

TG tok/s

658

TTFT (ms)

8.5

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.5

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,556	56.8	8.5 GB	current
4k	1,785	51.2	9.2 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	56.8	1.00×
2×	102.5	1.80×
4×	135.9	2.39×