Head to head
Llama Prompt Guard 2 86M vs LFM2.5 8B-A1B
On a blended workload of one million input tokens and three million output tokens, Llama Prompt Guard 2 86M is 2.4× cheaper than LFM2.5 8B-A1B.
| Llama Prompt Guard 2 86M | LFM2.5 8B-A1B | |
|---|---|---|
| Provider | Groq | Together AI |
| Input / MTok | $0.04 | $0.03 |
| Output / MTok | $0.04 | $0.12 |
| Cache read / MTok | — | — |
| Context window | — | 32,768 |
| Blended 1M in + 3M out | $0.16 | $0.39 |
| Status | preview | ga |