Context Compression Python Packages

distil-llm

Compression with a quality contract — cache-aware, causally-pruned LLM context compression for agentic runtimes, certified non-inferior across 7 domains. Works with any SDK.

entroly-core

Cut your Claude / OpenAI / Gemini bill 70–95% on AI coding. Local proxy that compresses context, keeps provider caches hot, and verifies LLM output ($0 hallucination guard). Drop-in for Cursor, Claude Code, Codex, Aider + 34 more and custom providers — 30s, no code changes

entroly

Cut your Claude / OpenAI / Gemini bill 70–95% on AI coding. Local proxy that compresses context, keeps provider caches hot, and verifies LLM output ($0 hallucination guard). Drop-in for Cursor, Claude Code, Codex, Aider + 34 more and custom providers — 30s, no code changes

llm-zip

Self-hosted HTTP sidecar for LLM context compression. Reduce token costs 3–5× before calling any AI API — powered by LLMLingua-2 and MarkItDown. No proxy, no API keys, no GPU required.

justllm

Production LLM calls. Just the three lines. Cross-provider fallback, native caching, and reversible context compression on by default.

claw-compactor

14-stage Fusion Pipeline for LLM token compression — reversible compression, AST-aware code analysis, intelligent content routing. Zero LLM inference cost. MIT licensed.

agent-coderag

CodeRAG: Lightweight semantic code search and distillation utility for AI coding agents. It solves the API knowledge gap via real-time local signature extraction and intent analysis. Optimized for token efficiency, it compresses codebase context into compact semantic summaries stored in a local DuckDB vector similarity index.

foldback-ai

Lossless context compression for LLM agents — fold it down, fold it back. Compresses tool outputs/logs before the model, prompt-cache-preserving, 39–82% fewer tokens. pip install foldback-ai

mnemosyne-engine

State aware knowledge compression, ingestion, and hybrid retrieval engine. Zero dependencies. Sub-100ms queries.

fish-bridge-mcp

An economical and fuel(token) efficient AI tool, graph session memory for looong chat session

foveance

Cut your LLM token bill 60%+ without changing your code or your answers. Drop-in proxy + one-line Python API that auto-compresses agent context for Claude Code, Codex, OpenAI, Anthropic & Ollama. pip install foveance

contextcrumb

Save Token Usage on Unstructured Document 😎. Let agent read docs, memories, prompts with in ultra-compressed mode through a tiny local model.

llm-context-compressor

Shrink LLM context windows by removing noise, redundancy, and long-tail detail — without losing signal.

tokenpack-rag

Query-aware semantic chunk selection under LLM context-window budgets.

supercompress

SuperCompress — learned context compression for LLMs.

mnemosyne-ollama

State aware knowledge compression, ingestion, and hybrid retrieval engine. Zero dependencies. Sub-100ms queries.

squeez

Squeeze verbose LLM agent tool output down to only the relevant lines

tooltrim

Drop-in compression for LLM agent tool outputs: cut ~99% of tool-result tokens while keeping (often improving) answer accuracy. Provider-agnostic, with a faithfulness benchmark and an OpenAI-compatible proxy.

mnemosyne-mcp

State aware knowledge compression, ingestion, and hybrid retrieval engine. Zero dependencies. Sub-100ms queries.

llama-index-postprocessor-vecr

LlamaIndex postprocessor for vecr-compress: deterministic retention-guaranteed node compression.

vecr-compress

Auditable, deterministic context compression for LLMs — structured data survives by regex whitelist, every decision logged.

langchain-vecr-compress

Deterministic, auditable LLM context compression — regex whitelist guarantees structured facts (IDs, URLs, dates, code) survive. Two layers: retention + heuristic knapsack.

compactbench

Open benchmark for LLM context compaction methods — measures what survives when you replace conversation history with a compacted artifact. Multi-cycle drift, hidden ranked set.