SuperCompress documentation

v2 is the hosted query-aware Neural Keep model: it keeps the context that answers the query and cuts the rest. Hosted API, and an agent plugin for coding agents.

⚡

Quickstart

Install, get a key, compress your first prompt, in under 5 minutes.

📖

API Reference

Python API, HTTP endpoints, response fields, error codes, and client usage.

🔌

Integrations

Wire SuperCompress into OpenAI, LangChain, Express.js, Vercel AI SDK, and more.

🚀

Usage

Call the hosted API, Neural Keep runs on the service, not your machines.

🧠

Architecture

How the learned policy works: features, training, inference, and compiler mode.

⚙️

Compress engine

Layman + technical notes for every function in compress-engine.js.

🌱

Environment & CO₂

Methodology for estimating energy and carbon savings from token reduction.

🤖

Coding Agents

Install the proxy, one command to compress context for Cursor, Windsurf, Claude Code, and more.

What is SuperCompress?

SuperCompress v2 is a query-aware context engine on the hosted API. It keeps what is relevant to answering the query and cuts the rest. Strongest on coding-agent dumps: tool output, logs, diffs, and stack traces.

It is strongest on coding-agent dumps: tool output, logs, diffs, and stack traces. Unlike truncation (which blindly drops the tail) or summarization (which rewrites IDs and errors), SuperCompress keeps the original lines that answer the current question. The query itself is never compressed.

Key facts

Metric	Value
Engine	Hosted Neural Keep (live) · query-aware
Quality metric	Evidence containment, required lines survive in original wording. See benchmarks.
Strongest on	Coding-agent dumps: logs, diffs, stack traces, tool output
Access	API and an agent plugin for coding agents
Hosted price	Free monthly allowance, then pay-as-you-go. Current plans are in the dashboard.
License	MIT, free for commercial use

When to use it

SuperCompress is designed for production AI features where context grows over time and token costs scale linearly with usage:

Chatbots & conversational AI, long chat histories where each turn appends to the context window
RAG pipelines, retrieved chunks that overshoot the context budget
AI search, indexing results, snippets, and document extracts sent with every query
Support copilots, customer transcripts, ticket history, knowledge base articles
Agent loops, tool call traces, intermediate outputs, log entries accumulating across turns
Memory systems, compressed episodic context for persistent agent memory

Not sure if it fits? Try the interactive playground, paste your actual context and see the compression results in real time. No signup needed.

How it works (briefly)

In compiler mode (the default), SuperCompress:

Segments the input context into semantic blocks (headings, code blocks, log entries, prose sections)
Scores each block against the user's query using a learned policy
Removes duplicate content, noise, boilerplate, and low-value blocks
Preserves context dependencies (nearby headings, imports, trace context)
Reports tokens saved, important context kept %, and compression risk level

The hosted Neural Keep path runs on the API (CPU today). A separate legacy compiler path exists for millisecond local preview, do not mix the two.