🦞 LLM Token Compression & Reduction Tool — Cut AI agent token costs by up to 97%. 6-layer deterministic context compression for AI agent workspaces. No LLM required. Prompt compression, context window optimization & cost reduction for any LLM pipeline.

The official repo for "LLoCo: Learning Long Contexts Offline"

Pytorch implementation for "Compressed Context Memory For Online Language Model Interaction" (ICLR'24)

🦞 龙虾饼 - 智能上下文压缩系统，让 AI 记忆永不溢出

Stop re-explaining your codebase to AI. Infinite speed memory + code graph for Claude Code & Codex CLI. 17 MCP tools, subagent protocol, hybrid search, TUI dashboard, crash recovery. Save 80-200K+ tokens/session.

Rolling context compression for Claude Code — never hit the context wall. Auto-compresses old messages while keeping recent context verbatim. Zero config, zero latency. Works as a Claude Code plugin.

Biological code organization system with 1,029+ production-ready snippets - 95% token reduction for Claude/GPT with AI-powered discovery & offline packs

Awesome list of paper on vision-based context compression

SYMB Session Intelligence | SYMB-Mem | context compression for Claude Code. All Data Is Important.

OpenClaw low-memory optimization guide for resource-constrained servers (2GB RAM)

天将｜LLM 智能体运行时资源控制引擎：Token 预算控制、成本优化、稳定性 Guard、多工具 Agent 调度

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

🌶️ OpenClaw skill that reduces token consumption by 30-60% through context compression and smart optimization

Exploring Context Compression techniques for token reduction. Fine-tuning LLMs for efficient text compression and reduced inference costs, analyzing the trade-offs with Q&A accuracy.

Agent memory runtime: short/long-term context, vector persistence, compression, and personalization primitives.

Exploring artificial compressed languages to improve efficiency, context usage, and cross-lingual unification in LLMs

Capture and compact long Google AI Studio sessions with multimodal image OCR and context handoff artifacts.

LLM context compression proxy — 40-70% token savings, zero code changes

Detecting silent pivot substitution in LLMs under context compression