A plugin for Claude Code, Codex, Copilot CLI and Pi that learns from past sessions in a repository to cut coding-agent token costs.
QuesmaOrg/awesome-ai-tokenomics
A curated awesome-list of tools, papers, benchmarks and vendor pages on what AI tokens cost, where they get wasted, and how to cut spend.
- Stars
- 215
- Forks
- 44
- Open issues
- 3
- Last commit
- yesterday
What is it?
Awesome AI Tokenomics is a curated list of resources on the economics of tokens used by LLMs and coding agents. It covers how token pricing works, how usage is measured, where spend leaks away, and how to reduce it.
The list is organized into four areas: Monitor (dashboards, tracing and OpenTelemetry instrumentation), Optimize (caching, compression, routing, gateways, context engineering and local models), Govern (budgets, chargeback, anomaly detection and spend management), and Measure (benchmarks, cache accounting and metering). Each entry is tagged by type, such as open-source tool, paper, benchmark, dataset or commercial product, with license and last-commit badges for GitHub-hosted projects.
The repository is maintained by Quesma and released under CC0-1.0, so its contents can be reused freely. It suits engineers and teams who want to understand or control the cost of coding agents and LLM applications.
Key features
- 01Covers monitoring tools for token usage and cost, including local dashboards for coding agents
- 02Lists optimization approaches such as prompt caching, context compression and model routing
- 03Includes governance resources like budget caps, chargeback and anomaly detection
- 04Provides papers, benchmarks and datasets for measuring token efficiency
- 05Tags each entry by type and license, with live last-commit badges for GitHub projects
Use cases
- Finding a tool to see which sessions or requests drive a coding agent's token bill
- Comparing context-compression or semantic-caching options before adding them to an LLM app
- Researching how vendors and academics measure token efficiency for agents
- Building a cost-control policy for a team that uses multiple LLM providers
In collections
Topics
FAQ
- What is awesome-ai-tokenomics?
- A curated awesome-list of tools, papers, benchmarks and vendor pages on what AI tokens cost, where they get wasted, and how to cut spend. Awesome AI Tokenomics is a curated list of resources on the economics of tokens used by LLMs and coding agents. It covers how token pricing works, how usage is measured, where spend leaks away, and how to reduce it.
- Is awesome-ai-tokenomics free to use? What license does it have?
- awesome-ai-tokenomics is open source under the CC0-1.0 license. Check the license terms before using it in commercial products.
- Is awesome-ai-tokenomics still maintained?
- Actively maintained. The last commit to awesome-ai-tokenomics was on Nov 6, 2026.
- What can I use awesome-ai-tokenomics for?
- Finding a tool to see which sessions or requests drive a coding agent's token bill Comparing context-compression or semantic-caching options before adding them to an LLM app Researching how vendors and academics measure token efficiency for agents Building a cost-control policy for a team that uses multiple LLM providers
- Where can I get awesome-ai-tokenomics?
- The source code and installation instructions are on GitHub: https://github.com/QuesmaOrg/awesome-ai-tokenomics
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A self-hosted LLM gateway for AI coding tools like Claude Code and Cursor that routes requests across providers and compresses tool output to cut token usage.
A Claude Code mod that keeps the prompt cache warm during breaks and estimates the rewrite cost before a cold send.
A community collection of custom agents, instructions, skills, hooks, workflows, and plugins for customizing GitHub Copilot.
A curated Chinese-language guide to ChatGPT and frontier LLMs, covering usage, prompting, app development, agents, and productivity tools.
A Rust binary that deterministically compresses coding-agent prompts and tool output, with removed content retrievable on demand.