Rank models on your own work.
Leaderboards score models on someone else's tasks. Stats reads your whole archive and draws your own Pareto frontier from what each conversation cost you and how it turned out.
Your own Pareto frontier.
Each model sits at its judged success on your conversations against its median cost, time, or output. The dotted line joins the models nothing beats on both. Give each reasoning effort its own dot to see whether high effort pays for itself on your work.
Compare modelsModel frontier
What you really pay.
List price bills every cached token as fresh input. Stats sets your median conversation with each model against that price, so the drop below the line is what caching saved you. Subscription value then says what each Claude and Codex plan is worth at API prices.
What a plan is worthHow far below list price do you pay?
Which models work on their own.
Each model sits at its tool calls and words per prompt you type, sized by how many conversations it ran, so agent-driven models stand apart from the ones you steer.
About Work & autonomyWhich models work most on their own?
How the work turns out.
A model you choose labels each finished conversation with its kind of work and how it ended, so Stats can show what goes well.
About classificationWhat kind of work is it?
- Build31%
- Fix22%
- Understand16%
- Plan9%
- Test8%
- Operate6%
- Analyze5%
- Orchestrate2%
- Communicate1%
How does it end?
- Succeeded61%
- Partially Succeeded19%
- Failed8%
- No clear goal12%
- Agents and chat apps
- 57
- Counted together, in one set of numbers.
- Uploads
- 0
- Stats are computed from the archive on your computer.
A graph of your projects.
Graph links conversations through shared files, commits, continuations, and agent messages, so related work turns up across harnesses and projects.
About the GraphYour year, in cards.
Year in review turns a year of conversations into cards you can save as images or share.
About Year in reviewMore of Hansard
Personal Sync
Keep your whole archive in storage you own, with every computer you use in step with the others.
Teams
Share whole conversations from the repositories you choose with your team, under rules every member can read.
Sharing
Send one conversation as a web page anyone can open, a PDF, Markdown, or a trajectory file.
Running and Review
Run each project's agents from one place, with tabs, changes, and terminals, and answer everything that waits on you in one queue.
Reading a run
Follow any agent run step by step: a trace of its calls, the agents it started, its context and cost, and Create PR.
Graph
See how your projects, conversations, files, commits, and decisions connect, and the evidence behind each link.
Slack bot
Start Claude Code and Codex runs on your Mac from Slack, approve each one in its thread, and get digests where you choose.
AI assistants
Let Claude.ai, ChatGPT, and other assistants search your agent conversations and memories, with your approval on your Mac.
Phone companion
Follow your agents from your phone, answer what they ask, and revive or start them while they run on your computer.
Accounts
Keep every Claude and Codex account in one place with its limits, route each conversation to one with allowance left, and cap API key spend.
Quality of life
Keep working when an account reaches its limit, and give every agent the MCP servers and skills you installed in one.
Put it on the record.
