The AI Gateway for coding assistants
Make your coding assistant faster, cheaper, and more reliable.
Compress tokens, route across models, and keep working, even when providers fail or limits are reached.
Free
For solo developers.
$0
- 1 developerThat's you!
- Connect all your coding agents (Claude Code, Codex, Copilot, OpenCode, Cursor…)Works wether your coding agent is used with a subscription or an api usage billing model
- Token compression (Input and Output)Reduce your token usage and costs by up to 50%
- Personal dashboard: usage, cost, sessionsView your usage, cost, and coding sessions in real-time
Team
For teams who wants more with their coding agent.
$29 / developer / month
Everything in Free, plus:
- BYOK + a daily allowance on our modelsBring your own token provider, plus tokens on our models for light-to-moderate usage
- Fallback ModelsKeep coding when a provider fails or rate-limits
- Reroute ModelsForced routing to specific models to save costs
- Team observability: manage developers and squads usageControl team-wide usage, coding agent access, and squad organization
- Spending cap per seatPrevent unexpected costs and get alerts when you're close to your limit
- GitHub integration (per repo / per PR attribution)Attribution helps you track usage and costs
Enterprise
For teams with scale, security, or compliance requirements.
Custom
Everything in Team, plus:
- On Demand Private OSS ModelsGet your own private OSS models hosted on Edgee. Pay per GPU hour for unlimited tokens.
- SSO / SAMLSingle Sign-On and Security Assertion Markup Language (SAML) for seamless authentication and access control.
- Private gateway (SaaS or on-prem)Get your own private gateway on Edgee or self-host it on your own infrastructure.
- Custom data residency (EU or US)Control where your data is stored and processed.
- Custom privacy controls (PII redaction, ZDR…)Add custom privacy controls to your prompts.
- Dedicated support & SLAGet a 30-minute SLA on Urgent tickets and 24/7 web and phone support via callback request.
Billing
How usage is billed
One simple rule: tokens you already pay for cost nothing extra through Edgee. You pay Edgee for seats and for the tokens it provides.
| Where your tokens come from | 🤖 Coding agentsCovered by the plans aboveClaude Code, Codex, Copilot, OpenCode, Cursor… | </> Apps & agentsPay as you go, no plan neededYour own products calling LLMs through Edgee's API. |
|---|---|---|
| Your own subscription or API keys (BYOK)You pay your provider directly | FreeNo markup. Fallback & rerouting included. | FreeNo markup. |
| Tokens provided by Edgee | IncludedToken pool with every Team & Enterprise seat. | 5% platform feeAdded when you top up Edgee credits |
| Private OSS models hosted by Edgee | Per GPU hourUnlimited tokens on your own dedicated models — talk to our team | |
Feature overview
Everything you need to build and scale
Core
- Edgee CLI
- Free
Gateway & Routing
- Multi-provider gateway
- Free
- All models
- Provider price (+5% on credit top-up)
- Fallback Models
- Team seat
- Reroute Models
- Team seat
Observability
- Debug logs
- Free
- Reports
- Free
Cost Controls
- Spending caps & alerts
- Team
Team & Usage
- Organization members beyond 1
- Team
- AI Gateway seat assignment
- Team
- GitHub integration
- Team
- Per-repo & per-PR attribution
- Team
- Extend usage beyond limits
- Team
Services
- Private models hosting
- Enterprise
- Private gateway
- Enterprise
Privacy & Admin
- Data policy routing
- Free
- SSO / SAML
- Enterprise
- Privacy controls (coming soon)
- Enterprise
Support
- Email support
- Free
- Priority support
- Team
- Contractual SLA
- Enterprise
Why Edgee
- Keep coding even when limits are reached or providers fail
- Reduce token usage automatically with built-in compression
- Route through multiple models with a single API
- Track usage per developer, repo, and team
- No lock-in — bring your own keys or use Edgee
Frequently asked questions
Start building without limits
Use Edgee for free — and upgrade when your team needs more control.
No credit card required
All plans include Edgee’s token compression. You save on LLM costs from day one.
