headroom
SHA-256Headroom compresses tool outputs, logs, RAG chunks, files, and conversation history before they reach the LLM. 60-95% fewer tokens with identical answers. Integrates as a library, proxy, MCP server, or agent wrapper.
Smart Download
Download Download Version
v0.22.4 · 15.3 MB
Compress everything your AI agent reads: 60-95% fewer tokens, same answers. Library, proxy, MCP, reversible.
Core Features
- Reversible compression (CCR): originals stored locally, LLM retrieves on demand
- 6 algorithms auto-selected: SmartCrusher (JSON), CodeCompressor (AST), Kompress-base (text), etc.
- Zero-code integration: agent wrapping (Claude Code/Codex/Cursor), inline library, transparent proxy
- Cross-agent memory: shared compressed context across Claude, Codex, Gemini, with auto-dedup
- Automated learning: mines failed sessions, writes corrections to AGENTS.md / CLAUDE.md
What It Can't Do
- •Requires Python 3.10+ and modern Node.js. Best suited for AI agent contexts; general text compression may be less dramatic. CCR needs local storage; if you can't keep files locally, disable CCR.
Use Cases
- Daily use with AI coding agents (Claude Code, Cursor, Aider) to slash API costs
- Multi-agent workflows where shared memory reduces redundant context transfer
Headroom is a context compression layer for AI agents that reduces token usage by 60-95% while preserving answer accuracy. It compresses everything your AI agent reads — tool outputs, logs, RAG chunks, files, and conversation history — before reaching the LLM. Unlike tools like LLMLingua or Selective Context, Headroom offers reversible compression (CCR) so originals are never lost, and cross-agent memory across Claude, Codex, Cursor, Aider, and more. It runs locally, supports 6 compression algorithms (SmartCrusher, CodeCompressor, Kompress-base, etc.), and integrates as a library, proxy, MCP server, or agent wrapper.
Tags
Getting Started
Download installer
Click the button above to download the installer for your system
Install the software
Open the downloaded dmg file, then drag the app to Applications
Install: pip install "headroom-ai[all]" or npm install headroom-ai
Choose mode: headroom wrap claude to wrap an agent, or headroom proxy --port 8787 for transparent proxy
Verify: headroom stats to see token savings
- Install: pip install "headroom-ai[all]" or npm install headroom-ai
- Choose mode: headroom wrap claude to wrap an agent, or headroom proxy --port 8787 for transparent proxy
- Verify: headroom stats to see token savings
SHA-256 checksum verified
Checksum extracted from GitHub official Release page
SHA256 Checksum
0942c5d45c6f94bb2cc7b094a323175bae262fdc3d38ab088e27aea6cc5186f3This checksum is extracted from the GitHub Release page. Verify file integrity after download.
All SHA-256 checksums on this platform are extracted from the project's official GitHub Release page, without any modification. You can independently verify them on the GitHub Releases page.
Open Source Transparency
View GitHub SourceUninstall Info
Uninstall the package: pip uninstall headroom-ai or npm uninstall headroom-ai. If you used `wrap`, remove ~/.headroom directory and any injected environment variables.
No Extra Dependencies
Ready to use after download. No additional runtime required.
Having issues? Check the FAQ below
5 FAQs
Similar Projects
ragflow
RAGFlow is a leading open-source RAG engine that fuses deep document understanding with agentic workflows, offering precise, citation-grounded answers. Unlike LangChain's DIY approach, RAGFlow provides a turnkey, enterprise-ready platform with a visual interface and template-based chunking, reducing hallucinations and setup complexity.
unsloth
Unsloth is the first desktop app to run and train AI models locally. It supports a wide range of models including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, and FLUX. Unlike most local model tools that focus only on inference, Unsloth adds a full training stack: fine-tune LLMs, diffusion, TTS, and embedding models 2× faster with 70% less VRAM. It also supports LoRA, QLoRA, RL, GRPO, DPO, and FP8. The app includes built-in Agents & Tools integration (Claude Code, Codex, MCP), private search and RAG, image/video diffusion, and audio support. With hardware support for CPU, NVIDIA, AMD, Intel, macOS, and multi-GPU setups, plus remote access via Cloudflare, Unsloth is a comprehensive local AI studio. It can export to GGUF, NVFP4, FP8 and offers an OpenAI-compatible API for easy deployment.
ollama
Ollama lets you download, run, and manage large language models locally. One command, multiple platforms, endless possibilities.