Head to head

LFM2.5 8B-A1B vs Gemma 3n E4B IT

On a blended workload of one million input tokens and three million output tokens, LFM2.5 8B-A1B is 1.1× cheaper than Gemma 3n E4B IT.

LFM2.5 8B-A1BGemma 3n E4B IT
Provider Together AI Together AI
Input / MTok $0.03 $0.06
Output / MTok $0.12 $0.12
Cache read / MTok
Context window 32,768 32,768
Blended 1M in + 3M out $0.39 $0.42
Status ga ga

LFM2.5 8B-A1B detail   Gemma 3n E4B IT detail