Edgee Team — for engineering teams
See every token. Across every dev. Across every repo.
Teams management gives engineering leaders full visibility into how their team uses Claude Code, Codex, and OpenCode. Track cost per repo and PR, manage team seats, and keep your team unblocked with automatic OSS model fallback.
Built on Edgee's token compression. Every call your team makes is automatically compressed — up to 50% token savings, on top of Team Management's features.
Tokens used this week
2.5M
Top repos
- acme/api-gateway1.2M$48.20
- acme/web-app842K$33.60
- acme/data-pipeline511K$20.40
Backed by founders and leaders of
Your team is shipping with AI. You have no idea what it costs.
Invisible spending
Anthropic and OpenAI bills land at month-end. By then, you've already burned your budget on an experimental repo no one's even shipping.
No attribution
Which dev uses the most tokens? Which repo eats your Max 20x seats? You can't tell until finance asks. And then it's too late.
Blocked teams
When a teammate hits a weekly limit mid-sprint, the whole team's velocity drops. And you find out on Slack, three hours later.
Team management
Manage your team in one place
Invite, manage, and control every dev on your team.
- Invite devs via email or SSO
- Assign roles — admin, member, read-only (perfect for finance and leadership)
- Set per-seat spending caps with automatic alerts at 50, 80, 100%
- Deactivate seats instantly when someone leaves. No lingering AI access.
Team members
6 of 10 seats
- AdminActive
Alex Chen
alex@acme.co
$312 / $500 - MemberActive
Sam Patel
sam@acme.co
$224 / $400 - MemberActive
Jordan Kim
jordan@acme.co
$368 / $400 - MemberActive
Morgan Silva
morgan@acme.co
$190 / $400 - Read-onlyActive
Riley Tan
riley@acme.co
— - MemberPending
Quinn Ortega
quinn@acme.co
$0 / $400
feat(auth): swap session store for Redis
Tokens in this PR
24,531
Direct cost
$0.82
GitHub integration
Attribute AI spend to the code it produces
Every token, tied to the code it helped ship.
- Connect your GitHub orgs in one click via OAuth
- See token consumption per repo, per PR, per branch, per file
- Drill down: which agent session produced which code
- Export usage as CSV, webhook, or REST API — plug into Looker, Tableau, or your internal BI
OSS fallback
Never let your team get blocked
When Claude's rate limit hits, when OpenAI has an outage, or when a dev exhausts their weekly cap, Team management automatically routes their requests to an open-source model. Coding doesn't stop.
- Configurable fallback chains: primary → secondary → tertiary
- Out of the box: GLM-5, Mistral Large, Llama 3.x, Qwen 2.5 Coder, DeepSeek Coder
- Self-hosted OSS models supported via your own edge endpoint (OpenAI-compatible)
- Transparent failover: your dev sees a notification in their terminal, not a crash
- Fallback usage tracked separately in your dashboard and billed at lower rates
Claude Code request
alex-chen · implement-feature.ts
Primary: Claude Opus
Weekly limit reached — 0 tokens remaining
Fallback 1: Mistral Large
Routed · 312ms to first token
Coding continues
Transparent to the dev
Dashboard & analytics
The dashboard your finance team will actually want
Total tokens (30d)
12.4M
+8% vs last 30d
Cost (30d)
$487
-22% vs last 30d
Compression savings
$412
9.6M tokens spared
Active devs
14 / 16
2 pending invites
Token usage (30 days)
Total across all devs and repos
Top 5 repos
- acme/api-gateway3.8M$152
- acme/web-app2.9M$116
- acme/data-pipeline2.1M$84
- acme/auth-service1.6M$64
- acme/analytics1.2M$48
Top 5 devs
- Alex Chen2.4M$96
- Sam Patel1.9M$76
- Jordan Kim1.5M$60
- Morgan Silva1.1M$44
- Riley Tan0.8M$32
Real-time spend
See usage and cost as it happens. Not at month-end.
Savings report
Exact tokens saved by Edgee compression, in dollars and in extended sessions.
Budget alerts
Thresholds per dev, per repo, per team. Slack and email notifications.
Export everything
CSV, webhooks, REST API. Plug it into Looker, Tableau, or your BI of choice.
Simple to set up. Privacy-first by design.
- 1
Install
Your devs run `edgee launch claude` as usual. Nothing changes in their local workflow.
- 2
Connect
You create your Team workspace, invite devs, and connect GitHub. Under 30 minutes for a team of 20.
- 3
Observe
Every call is compressed, routed, tracked, and attributed automatically.
Your code never touches our servers. Team management stores metadata only — token counts, timestamps, repo names, compression ratios. Prompts transit the edge but are never persisted.
Simple pricing that scales with your team.
Team
Billed annually · Volume discounts from 20 seats
- Unlimited Edgee token compression
- Team management (up to 100 seats)
- GitHub integration (unlimited repos)
- OSS model fallback
- Full team dashboard + exports
- Slack & email alerts
- Email support
Enterprise
For teams of 100+ or with specific compliance needs
- Everything in Team, plus:
- SSO & SAML
- Audit logs
- Self-hosted option
- Dedicated Slack channel + 99.9% SLA
- Custom data residency (EU, US, or private)
- Volume-based custom pricing
All plans include Edgee's token compression. You save on LLM costs from day one.
Edgee Free vs Team
Everything you already get on the Free plan, plus team-level features your finance and leadership actually use.
| Plan | Free | Team |
|---|---|---|
| Token compression | ||
| Individual dashboard | ||
| Multi-agent support (Claude Code, Codex, OpenCode) | ||
| Team management | ||
| GitHub integration | ||
| Per-repo and per-PR attribution | ||
| OSS model fallback | ||
| Spending caps and alerts | ||
| SSO / SAML | Enterprise | |
| Self-hosted | Enterprise | |
| Priority support |
Frequently asked questions
Stop flying blind on AI coding spend.
See your team's AI usage, save tokens, and ship faster. 14 days free.
No credit card required. Cancel anytime during trial.