← Back to community benchmarks

Llama-3.2-1B-Instruct

M2 Max (38c) · 32 GB · 4bit · 2026-03-14
Performance
1k
tokens
2,946
PP tok/s
296.3
TG tok/s
348
TTFT (ms)
1.2
Peak mem (GB)
Hardware
Chip M2 Max (38c)
Memory 32 GB
GPU Cores 38
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 2,946 296.3 1.2 GB current
4k 3,781 219.2 1.5 GB view
Batching Results
Batch Size TG tok/s Speedup
296.3 1.00×
464.8 1.57×
426.1 1.44×