Headroom
Latest release v0.37.0 on 2026-08-27 · Actively maintained · 15 releases in 90 days
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
| Latest release | v0.37.0 on 2026-08-27 |
|---|---|
| Maintenance | Actively maintained — pushed within the last 14 days, with 3 or more releases in 90 days |
| Stars | 73,355 |
| Releases in 90 days | 15 |
| Last commit | 2026-09-19 |
| Repository | headroomlabs-ai/headroom |
| Licence | Apache-2.0 |
| Open issues | 706 |
Recent releases
v0.37.0 2026-08-27
Fixed
- Cache/semantic: don't semantic-match an empty query across contexts (455f4f2)
- Copilot: preserve native enterprise model routing (997a479)
- Kimi: route managed Kimi Code through the proxy (701e461)
- Learn: surface Codex analysis failures (632cb81)
v0.36.5 2026-08-22
Fixed
- Codex: detect ChatGPT auth from idtoken claims so wrap/init emit requiresopenaiauth (2f81fa5)
- Doctor: report project-scoped Claude routing instead of a false negative (8f3e33a)
v0.36.4 2026-08-22
Fixed
- Dashboard: pin MIME types for the vendored static assets (b485768)
- Proxy/responses: keep the Codex additionaltools carrier on the wire (1617f83)
- Security: validate caller-supplied upstreams on every resolution path (3e3c409)
- Skip cross-turn dedup pointers on OpenAI chat streaming (9c30b62)
Maintenance status is derived from public repository activity — release cadence and commit recency. It is not a review, an endorsement, or a security assessment. Star counts measure attention, not quality.