clocktowerbench.com
clocktowerbench.com is You can see the full benchmark at the website, and I’ve also made it so that you and your friends can play again the AIs. It is ranked #11 on the AI Radar, in Frontier models, first seen 15 h ago and shared in 19 posts (559.8k views).
What clocktowerbench.com says about itself
Frontier AI models play Blood on the Clocktower against each other. One model plays the town, another the Demon's team. Ratings, every game's log, and a table where you can play them.
Clocktower Bench: AI models playing Blood on the Clocktower Skip to content Clocktower Bench Clocktower Bench AI models play the social deduction game Blood on the Clocktower against each other: one model plays the town, another the Demon's team, hiding among them. Who catches liars, and who lies best? Video Watch the video essay What happens when AIs play a game of lies. Essay Read the write-up The findings in full, on Substack. Games Browse every game 538 replays with every seat's reasoning. Play Play against the AIs Take a seat at a table of models. Leaderboard Every model plays both sides. Its rating combines how well it hunts the Demon…
What people said about clocktowerbench.com on X
I spent 20,000 dollars in API credits having different AI models play each other at Blood on the Clocktower (a Mafia or Werewolf type social deception game). The results were fascinating. Models have gotten much better over the course of this year at creative on-the-spot scheming and bluffing, and also at reasoning…
— @dwarkesh_sp, 15 h ago · 2.3k likes · see the post
Honestly quite impressed with the methodology on this benchmark. @dwarkesh_sp want a job designing evals at @OpenAI?
— @polynoamial, 11 h ago · 1.7k likes · see the post
OpenAI models demolishing Anthropic models at Blood on the Clocktower (Astra has a 82% win rate against Opus 5.5).
— @dwarkesh_sp, 13 h ago · 446 likes · see the post
Alternatives to clocktowerbench.com
- Thinking Machines Lab — Mantic have been using Tinker since it launched. This guest post is a technical deep dive on what they have built so…
- openai.com —
- Tencent Hy — Tencent Hy
- Claude — Basecamp for Claude is live!
- Mercor — Evaluate frontier AI models with Mercor's APEX AI Benchmarks & Leaderboards. Measure AI productivity across software…
- openai.com —
clocktowerbench.com in numbers
- Rank on the AI Radar: #11 of 10384
- Shared in 19 posts by 17 accounts: @dwarkesh_sp, @polynoamial, @teortaxesTex, @lugaricano
- 559.8k views on those posts
- First seen 15 h ago, last shared 1 h ago
- Pricing seen by AGK Intelligence: free
- Market: Frontier models
FAQ
What is clocktowerbench.com?
You can see the full benchmark at the website, and I’ve also made it so that you and your friends can play again the AIs It was first shared on X 15 h ago and is ranked #11 on the AI Radar.
Is clocktowerbench.com free?
Yes, it is free to use.
Who shared clocktowerbench.com?
17 accounts on X, including @dwarkesh_sp, @polynoamial, @teortaxesTex, in 19 posts totalling 559.8k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.7k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 12:45 UTC. Full method.