← Back to community benchmarks

Llama-3.2-3B-Instruct

M4 (8c) · 16 GB · 4bit · 2026-03-09
Performance
4k
tokens
382.4
PP tok/s
37.8
TG tok/s
10715
TTFT (ms)
2.6
Peak mem (GB)
Hardware
Chip M4 (8c)
Memory 16 GB
GPU Cores 8
Software
oMLX v0.2.6
macOS macOS 15.7.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 419.8 47.9 2.3 GB view
4k 382.4 37.8 2.6 GB current
Batching Results
Batch Size TG tok/s Speedup
47.9 1.00×
84.8 1.77×
88.5 1.85×