llama-swap

What you can use it for

Switch local model servers behind a compatible API.

• Serve and automatically switch local models behind a compatible API instead of a managed gateway for local agent traffic. Not a full cloud-provider gateway replacement; hardware and model-serving setup are required.

Before you switch

Local-serving slice only: needs llama.cpp or another compatible model server, configuration and enough RAM/VRAM. Model licences vary. Does not reproduce managed cloud-provider billing, credit purchasing or all gateway analytics.

Alternatives to these paid apps

Each comparison covers the workflow described above, rather than every paid feature.

LLM Gateway→Website ↗