Head to head

LFM2.5 8B-A1B vs Ministral 3 8B

On a blended workload of one million input tokens and three million output tokens, LFM2.5 8B-A1B is 1.5× cheaper than Ministral 3 8B.

LFM2.5 8B-A1BMinistral 3 8B
Provider Together AI Mistral
Input / MTok $0.03 $0.15
Output / MTok $0.12 $0.15
Cache read / MTok
Context window 32,768
Blended 1M in + 3M out $0.39 $0.60
Status ga ga

LFM2.5 8B-A1B detail   Ministral 3 8B detail