Headroom for Teams
Headroom Enterprise for engineering teams — more savings, automatic model routing, attack removal, SSO and an org-wide savings dashboard. 3 months free with a 12-month plan.
Headroom OSS is built for one developer on one laptop. Headroom Enterprise is built for the whole team.
| Open source | Enterprise | |
|---|---|---|
| Savings | Compression, plus tool search on Anthropic requests (on by default) | Also: tool search for every model, harness tuning, extra compressors |
| Model routing | Rule-based and opt-in: turn it on with HEADROOM_MODEL_ROUTER_ENABLED=1 and write the rules in HEADROOM_MODEL_ROUTES | Automatic: a cheaper model per request when quality allows, with a shadow mode that shows the savings before you turn it on |
| Security | — | Detect and strip prompt injection and risky tool calls |
| Rollout | Per laptop | Managed or in your VPC, SSO, central config, org-wide savings |
| Support | GitHub, Discord | Dedicated engineers, onboarding, priority response |
3 months free
Companies that sign a 12-month Enterprise plan get their first 3 months free. One perk per company.
Prefer to talk first? Email hello@headroomlabs.ai with your stack and rough monthly LLM spend.
Everything in the open-source repo stays open source under Apache 2.0.
Persistent Installs
Install Headroom as a durable local runtime — background service, scheduled watchdog, or restartable Docker container — instead of starting it ad hoc.
How Compression Works
Understand Headroom's compression pipeline, automatic content routing, and how different content types are compressed.