← Back to community benchmarks

Meta-Llama-3.1-8B-Instruct

M4 (8c) · 16 GB · 4bit · 2026-03-10
Performance
1k
tokens
168.3
PP tok/s
21.7
TG tok/s
6091
TTFT (ms)
4.9
Peak mem (GB)
Hardware
Chip M4 (8c)
Memory 16 GB
GPU Cores 8
Software
oMLX v0.2.7
macOS macOS 26.3
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 168.3 21.7 4.9 GB current
4k 160.1 19.9 5.3 GB view
Batching Results
Batch Size TG tok/s Speedup
21.7 1.00×
38.3 1.76×
37.5 1.73×