Head to head
Llama 3.3 70B Versatile vs Gemma 4 31B IT
On a blended workload of one million input tokens and three million output tokens, Llama 3.3 70B Versatile is 1.1× cheaper than Gemma 4 31B IT.
| Llama 3.3 70B Versatile | Gemma 4 31B IT | |
|---|---|---|
| Provider | Groq | Together AI |
| Input / MTok | $0.59 | $0.39 |
| Output / MTok | $0.79 | $0.97 |
| Cache read / MTok | — | — |
| Context window | 131,072 | 262,144 |
| Blended 1M in + 3M out | $2.96 | $3.30 |
| Status | ga | ga |