Infercom is Infere's own inference provider, serving open-weight models such as Gemma, gpt-oss, Llama, and DeepSeek classes alongside third-party providers. Because it appears in the same catalog, Infercom models can be routed to, fall back to, and budgeted against exactly like any other provider.
Why it matters
First-party inference gives Infere another price and capacity option for open-weight models, without a separate integration.
How it works
Infercom serves open-weight models alongside third-party providers and appears in the same catalog. That means its models can be routed to, used as fallbacks, and budgeted against exactly like any other provider.