← Back to community benchmarks

Llama-3.2-1B-Instruct

M5 (10c) · 24 GB · 4bit · 2026-03-13
Performance
1k
tokens
3,183
PP tok/s
135.8
TG tok/s
322
TTFT (ms)
1.3
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.2.10
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 3,183 135.8 1.3 GB current
4k 4,604 100.2 1.4 GB view
Batching Results
Batch Size TG tok/s Speedup
135.8 1.00×
261.8 1.93×
319.3 2.35×