Find tools › Open-source projects › Spend less on AI

billion-context
Open-source project · by ranxianglei
Shrinks long chat histories as it goes, so AI coding helpers can work longer for less money.
Some setupWell known3,520,997 installs a month
What it costs
Free and open source; you may pay for the AI model it uses.
Use it to
- Shrink token bills on multi-day coding sessions
- Fit long agent conversations into smaller context windows
- Maintain prefix cache rates while compressing old messages
How to start
Install globally using
npm install -g billion-context --prefix=~/.local.Launch your preferred coding agent through the proxy CLI (such as
bili opencodeorbili pi).Point your agent base URL to the local billion-context proxy address.
Before you install
- Found automaticallyNot reviewed by a person yet. Check it before installing.
- Last updated 2026-11-18
More detail
An API proxy and CLI tool that sits between coding agents and model providers to dynamically compress conversation history while preserving cache hit rates.
Long agent sessions hit token limits and cost significant money; this lets sessions run across millions of tokens by letting the model compress, search, and decompress earlier conversation ranges.
Similar tools
Descriptions are written in plain English from each project's own documentation and refreshed by our agent; this one was last checked 2026-10-10. Spotted a mistake? · Open in Find tools · © 2026 Which AI Map