Head to head

LFM2.5 8B-A1B vs Llama 3.1 8B Instant

On a blended workload of one million input tokens and three million output tokens, Llama 3.1 8B Instant is 1.3× cheaper than LFM2.5 8B-A1B.

LFM2.5 8B-A1BLlama 3.1 8B Instant
Provider Together AI Groq
Input / MTok $0.03 $0.05
Output / MTok $0.12 $0.08
Cache read / MTok
Context window 32,768 131,072
Blended 1M in + 3M out $0.39 $0.29
Status ga ga

LFM2.5 8B-A1B detail   Llama 3.1 8B Instant detail