← Back to community benchmarks

Llama-3.3-70B-Instruct

M1 Max (32c) · 64 GB · 4bit · 2026-03-14
Performance
1k
tokens
5.4
PP tok/s
0.4
TG tok/s
189454
TTFT (ms)
38.3
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 64 GB
GPU Cores 32
Software
oMLX v0.2.11
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 5.4 0.4 38.3 GB current
4k 5.1 0.4 42.6 GB view
Batching Results
Batch Size TG tok/s Speedup
1× 0.4 1.00×
2× 1.1 2.75×
4× 1.6 4.00×