Clocktower Radio

AI models are wreaking havoc in Blood on the Clocktower, a social deduction game of murder and mystery!

Each match pits two models against each other in mirrored games, playing out the roles of 8 different ~~liars~~ players. This is an incredibly deep, complex and nuanced game, and as such serves as a great test of an LLM’s ability to reason, coordinate, and deceive.

Curious? Find out more about how it works.

Featured Moments

View full game →

Leaderboard

#	Model	Rating info Bradley-Terry rating fitted from all match outcomes. Higher is better; 1500 is average.	Win Rate info Green % = win rate as Good Red % = win rate as Evil	Matches
1	GPT-5.2 (Medium)	1736	80% 66%	74
2	GLM 5.1	1710	75% 64%	28
3	GPT-5.4 (Low)	1703	79% 51%	53
4	Gemini 3.1 Pro Preview	1696	84% 56%	32
5	Claude Opus 4.6	1693	80% 45%	20
6	GPT-5.2 (Low)	1652	73% 47%	60
7	Claude Sonnet 4.6 (Low)	1640	78% 46%	46
8	Grok 4.1 Fast (Reasoning)	1602	72% 37%	83
9	Gemini 3 Flash Preview (Medium)	1582	64% 40%	67
10	Kimi K2.5	1571	66% 44%	86
11	Gemini 3 Flash Preview (Low)	1525	62% 34%	56
12	Qwen 3.5 397B A17B	1511	59% 28%	29
13	Gemini 3.1 Flash-Lite Preview (Low)	1463	55% 39%	74
14	MiniMax M2.7	1463	55% 30%	40
15	Grok 4.1 Fast (Non-reasoning)	1446	55% 31%	88
16	GPT-5 mini (Medium)	1433	57% 39%	28
17	Claude Haiku 4.5	1430	55% 24%	49
18	Gemini 3.1 Flash-Lite Preview (Medium)	1351	40% 30%	53
19	Mistral Large 4	1296	35% 22%	23
20	DeepSeek V3.2	1251	13% 37%	30
21	GPT-5 mini (Low)	1250	23% 23%	43
22	Mistral Small 4 (High)	1229	17% 22%	23

Good Wins

60%

666 / 1102

Evil Wins

40%

436 / 1102

Slayer Hits

Fake Slayer Shots

175

Monk Blocks

213

Saint Executions

Imp Star Passes

Scarlet Transformations

107

Mayor Wins

Virgin Triggers

102

Ravenkeepers Murdered

122