Hacker GitHub

← All repos

headroomlabs-ai/headroom

View on GitHub ↗ ★ 78,618 Python

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.

26 mentions on Hacker News by 20 people: 12 stories, 14 comments · first 2026-01-13 · latest 2026-09-04

Mentions per year 26 mentions · busiest year 2026 (26)
20082026
AllStoriesComments NewestTopOldest
  1. How is this different from headroom? Mcptoon seems like it's specific to tool calls. https://github.com/headroomlabs-ai/headroom
    comment by swedishagentic · 2026-08-11 · in Show HN: Mcptoon – Token-efficient MCP CLI client
  2. …They provide perfect GLM5.2, awesome token/s, TTFT and price. Coupled with a local Headroom ( https://github.com/headroomlabs-ai/headroom ) you'll be able to use a LOT without hitting your 5h window :) Definitely the best $ value for me…
    comment by w4yai · 2026-06-22 · in GLM 5.2 vs. Opus
  3. …types. I liked their approach a lot better than RTK and thought it might be relevant for you. https://github.com/chopratejas/headroom
    comment by jvican · 2026-06-19 · in The Token Compression Illusion: Why I'm Skeptical of RTK
  4. Anybody have experience with https://github.com/chopratejas/headroom ? They seem to have similar goals in token reduction, but headroom appears to be broader in scope.
    comment by akman · 2026-06-19 · in The Token Compression Illusion: Why I'm Skeptical of RTK
  5. …built-in LSP support. I point it at the tailnet DNS (https://llama.<tailnet>:<port>) - Headroom ( https://github.com/chopratejas/headroom ) to maximize the 80k context window - Exa MCP for web search ( https://exa.ai/ ) this alone makes…
    comment by porkloin · 2026-06-15 · in Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?
  6. The github seems to be a better source, since it supports JavaScript and python https://github.com/chopratejas/headroom (4 points, 16 days ago) https://news.ycombinator.com/item?id=48155888 But we also seem to have a couple of Show HN from…
    comment by gnabgib · 2026-05-31 · in Headroom compresses everything your AI agent reads before it reaches the LLM
  7. …ago: https://news.ycombinator.com/item?id=47914963 Headroom:The Context Optimization Layer for LLM: https://github.com/chopratejas/headroom Of these, it looks like RTK and 8v are somewhat equivalent, while headroom is complementary.…
    comment by esperent · 2026-05-03 · in Context Mode v1.0.0
  8. …which I guess is a close equivalent. https://github.com/rtk-ai/rtk Headroom which is complementary. https://github.com/chopratejas/headroom I have tried both of these and I have to say, I didn't notice any difference. That's not to say…
    comment by esperent · 2026-04-27 · in 8v: One CLI for you and your AI agent. Up to 66% fewer tokens
  9. …Headroom out and looks like Headroom actually uses RTK under the hood for CLI output compression: https://github.com/chopratejas/headroom#compared-to
    comment by slawr1805 · 2026-04-20 · in Claude Token Counter, now with model comparisons
  10. …RTK - it includes RTK sure, but headroom cli has a lot more going on under the hood. For more see https://github.com/chopratejas/headroom
    comment by gghootch · 2026-04-17 · in Claude Opus 4.7
  11. …you want to reduce token usage is headroom https://github.com/gglucass/headroom-desktop (mac app) https://github.com/chopratejas/headroom (cli)
    comment by gghootch · 2026-04-16 · in Claude Opus 4.7
  12. …Some great ones are https://github.com/gglucass/headroom-desktop https://github.com/rtk-ai/rtk https://github.com/chopratejas/headroom https://github.com/samuelfaj/distill
    comment by gghootch · 2026-04-08 · in Sonnet 4.6 Elevated Rate of Errors
  13. …compresses context between you and Claude by ~34%. Sits at localhost, zero config once running. https://github.com/chopratejas/headroom > RTK - Rust CLI proxy that compresses shell output (git, npm, build logs) by 60-90% before it hits the…
    comment by sillysaurusx · 2026-03-31 · in Universal Claude.md – cut Claude output tokens
  14. …"headroom-ai[proxy]" && headroom proxy --port 8787 then point your client’s base URL at it. Repo: https://github.com/chopratejas/headroom Limitations: best on JSON arrays; text compression is opt-in; if you truly need every row, you’ll…
    comment by chopratejas · 2026-01-13 · in Headroom – context optimization layer for tool-using agents