repoGitHubTrust 82 · PrimaryPublished 1mo agoLive · yesterday
headroomlabs-ai/headroom
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.
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 50%I spent ~4.5 months building a free, self-hosted AI gateway: one endpoint for 237 providers (90+ free), auto-fallback, and a token-compression pipeline (MIT) →
- PossiblePossibly related (embedding) · 46%[R] Compiling Agentic Workflows into LLM Weights: Near-Frontier Quality at Two Orders of Magnitude Less Cost →
- LinkedLinked via github contributor · 85%chopratejas →
“995 commits”
- LinkedLinked via github contributor · 85%JerrettDavis →
“265 commits”
- LinkedLinked via github contributor · 85%gglucass →
“83 commits”
- LinkedLinked via github contributor · 85%adryanev →
“34 commits”
- LinkedLinked via github contributor · 85%MrAshRhodes →
“30 commits”
- LinkedLinked via github contributor · 85%DevanshiVyas →
“20 commits”
Covers
contributed_to (incoming)
personchopratejaspersonJerrettDavispersongglucasspersonadryanevpersonMrAshRhodespersonDevanshiVyaspersonagaonkerpersonrodboevpersonParideboypersonKayzopersondependabot[bot]personmbachaudpersonpratikbinpersonneogenixpersonaivinaypersonEyalm321persongithub-actions[bot]personinix-xpersonousamabenyounespersonRTCartistpersonlifeodysseypersonLakshya77089persontenderdevepersonangptpersonehendrix23
Related to (incoming)
Related across the graph
personParideboypersonpratikbinpaperTraceLab: Characterizing Coding Agent Workloads for LLM ServingpersonLakshya77089personagaonkerpersonEyalm321persongglucasspersonangptpersonehendrix23personJerrettDavispersonRTCartistpersonneogenixpersonadryanevpersongithub-actions[bot]personMrAshRhodespersondependabot[bot]personlifeodysseypersonmbachaudpersonchopratejaspersoninix-xpersonDevanshiVyaspersonaivinaypersonrodboevpersonKayzonewsI spent ~4.5 months building a free, self-hosted AI gateway: one endpoint for 237 providers (90+ free), auto-fallback, and a token-compression pipeline (MIT)personousamabenyounespersontenderdevenews[R] Compiling Agentic Workflows into LLM Weights: Near-Frontier Quality at Two Orders of Magnitude Less Cost