← Back to community benchmarks

LFM2.5-8B-A1B

M5 Pro (20c) · 64 GB · 8bit · 2026-07-05
Performance
1k
tokens
2,956
PP tok/s
143.9
TG tok/s
346
TTFT (ms)
8.9
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.4.4
macOS macOS 26.5.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,956 143.9 8.9 GB current
4k 3,905 141.3 9.3 GB view
Batching Results
Batch Size TG tok/s Speedup
143.9 1.00×
170.6 1.19×
222.5 1.55×