Edgee

Edgee, from Edgee Cloud, is an intelligent routing layer that sits between AI agents and language model providers, optimizing token spending through smart routing, compression, and cost tracking, claiming up to 70% cost reduction. Smart routing and budgets provide budget-driven routing to the right model at the right price with per-developer or team-level controls, token compression reduces token counts through tool result trimming without requiring code changes, and team observability tracks spending and usage by developer, repository, and pull request. Open weight model access provides GLM, Kimi, and MiniMax at flat monthly rates, bring-your-own-keys lets teams use their own provider credentials, and on-premise deployment supports self-hosting via Docker or Kubernetes for air-gapped environments.