Edgee is an AI gateway designed to sit between your agents and LLM providers, compressing prompts to reduce token costs and extend context windows. It serves developers and teams using coding agents or building applications, providing a transparent proxy that requires no code changes to integrate.
In the AI-driven technological shift, intelligence has a cost measured in tokens, which flow through every interaction and decision. Edgee addresses the problem of escalating token expenses and provider reliability issues, enabling more efficient and cost-effective AI operations without altering application code.
Edgee applies token compression through two layers: Layer 1 for input tool-result trimming and Layer 2 for output brevity. This process cuts tool-result payloads by 60–90% at the edge while remaining semantically lossless for coding tasks, delivering the same model output with fewer tokens billed.
The gateway offers intelligent routing with automatic fallback, retrying failed requests and switching to the next available provider transparently. This ensures uninterrupted service and reliability for coding sessions and applications without requiring manual intervention.
Team management features provide full visibility into team usage of coding agents, tracking cost per repository and pull request, managing team seats, and keeping teams unblocked. Observability tools monitor latency, errors, usage, and cost per model, app, and environment.
Edgee works as a transparent proxy, applying three pillars to every request: compress, route, and observe. It installs in under a minute via CLI, connecting to coding agents like Claude Code, Codex, Copilot, OpenCode, and Cursor, instantly applying compression and routing logic.
Users benefit from up to 50% cost reduction, over 30% longer coding sessions due to extended context windows, and enhanced reliability with automatic fallback. Teams gain cost governance, usage insights, and seamless integration without code modifications.
Use cases include integrating with coding agents to save tokens on development tasks, building applications that leverage multiple LLMs with a single API, and managing team AI usage and costs across repositories and projects. It supports any agent or app requiring LLM interactions.
admin
Target users are developers, engineering teams, and organizations using AI coding assistants or building LLM-powered applications. Integrations work with coding agents like Claude Code, Codex, Copilot, OpenCode, and Cursor, and it supports Bring Your Own Keys for billing control. The tech stack includes a CLI for installation and an OpenAI-compatible API gateway.
Edgee enables efficient intelligence movement by compressing, routing, and optimizing token flow, cutting costs and extending capabilities for AI-driven workflows without code changes.
Edgee targets developers, engineering teams, and organizations using AI coding assistants like Claude Code, Codex, Copilot, OpenCode, and Cursor, or building LLM-powered applications. It is designed for those seeking to reduce token costs, extend context windows, and manage AI usage without code changes, offering tools for cost governance, reliability, and observability.