Head to head
Llama Prompt Guard 2 22M vs Gemma 3n E4B IT
On a blended workload of one million input tokens and three million output tokens, Llama Prompt Guard 2 22M is 3.5× cheaper than Gemma 3n E4B IT.
| Llama Prompt Guard 2 22M | Gemma 3n E4B IT | |
|---|---|---|
| Provider | Groq | Together AI |
| Input / MTok | $0.03 | $0.06 |
| Output / MTok | $0.03 | $0.12 |
| Cache read / MTok | — | — |
| Context window | — | 32,768 |
| Blended 1M in + 3M out | $0.12 | $0.42 |
| Status | preview | ga |