← Back to community benchmarks

Llama-3.2-1B-Instruct

M1 (8c) · 8 GB · 4bit · 2026-03-09
Performance
1k
tokens
823.0
PP tok/s
65.9
TG tok/s
1245
TTFT (ms)
1.3
Peak mem (GB)
Hardware
Chip M1 (8c)
Memory 8 GB
GPU Cores 8
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 823.0 65.9 1.3 GB current
4k 800.9 45.2 1.4 GB view
Batching Results
Batch Size TG tok/s Speedup
65.9 1.00×
110.4 1.68×
117.9 1.79×