← Back to community benchmarks

Qwen2.5-3B-Instruct

M4 (10c) · 16 GB · 6bit · 2026-03-10
Performance
8k
tokens
482.3
PP tok/s
30.5
TG tok/s
16987
TTFT (ms)
3.1
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.6
macOS macOS 26.0.1
Context 8,192