Agent gateway for engineering teams: route coding-agent traffic across 200+ models, compress tokens, and track spend.
Edgee homepage
1 / 3Overview
What is Edgee?
Take back control of your token spend.
What it is
Edgee is an agent gateway for engineering teams that run AI coding assistants. It sits between your agents and model providers as a unified API gateway: routing requests across 200+ models, compressing tokens, and showing exactly where every token goes — with a stated goal of up to 70% cost reduction.
Why teams pick it
AI coding tools are powerful but their spend is hard to see and harder to control. Edgee puts every coding agent on a budget you set, routes each request to the right model at the right price, and compresses what remains. It works with your existing Claude, Codex, or Cursor subscriptions or your own API keys, with tokens billed at provider prices and no markup.
Key capabilities
Budget-driven routing sends each request to the best-value model, with custom rerouting strategies or automatic smart routing by task complexity. Input/output compression cuts token usage for the same context. A real-time dashboard shows requests, tokens, cache hits, and per-session savings, with shareable reports. The Edgee Mod for Claude Code adds an interactive sidebar inside the terminal: switch models in one keystroke and watch savings live, with no separate plugin to install.
Pricing and access
The Free plan ($0) covers one developer with unlimited coding agents, unlimited tokens, rerouting strategies, and compression — no credit card required. Team is $30/developer/month with SSO, spend controls, and smart routing beta. Enterprise adds private on-prem gateways and custom privacy controls. Install takes two commands on macOS, Linux, or Windows.
Platforms and languages
Edgee Availability
Platforms
Languages
Capabilities
Edgee Key Features
Claude Code Mod
Switch Claude Code sessions to any of 200+ models in one keystroke and see live cost savings, right inside the terminal.
Budget-driven model routing
A unified gateway that routes each request to the best-value model under budgets you define; smart routing picks by task complexity.
Token compression
Shrinks input and output tokens for the same context; compression and routing savings are tracked per session.
Real-time cost dashboard
Requests, tokens, cache hits, and savings per session, with shareable session reports for the team.
BYOK and easy install
Works with existing Claude, Codex, Cursor subscriptions or your own keys at provider prices, no markup; two-command install.
Best for
Who uses Edgee?
Developers and engineering teams using AI coding assistants (Claude Code, Codex, Cursor, etc.)
Cut model token spend and manage multi-model usage under team budgets
Plans and access
Edgee Pricing
Freemium
Free: 1 developer, unlimited agents/tokens. Team $30/dev/mo. Enterprise custom.
Common questions
Edgee FAQs
Do I need a credit card to start?
No. The Free plan starts without a credit card, and you can also connect your existing provider subscriptions.
What does the Free plan include?
One developer with all their coding agents, rerouting strategies, and input/output compression — unlimited tokens.
Rerouting Strategies vs Smart routing?
Rerouting Strategies are budget rules you define yourself (available on all plans); Smart routing automatically picks a model by task complexity and is currently a Team/Enterprise beta.
Does the Claude Code Mod need a separate plugin?
No. It ships with the Edgee CLI — launching Claude Code through Edgee loads it automatically and replaces the old statusline display.
Which models can the Mod switch between?
Every model your current coding-assistant key can reach, including available open-weight models.
Reviews
See what the community thinks and share your experience.
—
Based on 0 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
