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. The ± shows the margin of error (or 95% CI).	Win Rate info Green % = win rate as Good Red % = win rate as Evil	Matches
1	Kimi K2.6	1793±108	81% 78%	32
2	Gemini 3.1 Pro Preview	1709±91	86% 56%	36
3	GPT-5.2 (Medium)	1701±63	76% 64%	78
4	GLM 5.1	1695±68	76% 61%	41
5	GPT-5.4 (Low)	1687±59	77% 53%	57
6	Claude Opus 4.6	1650±82	68% 46%	28
7	Claude Sonnet 4.6 (Low)	1621±54	73% 47%	49
8	GPT-5.2 (Low)	1619±58	70% 44%	63
9	Grok 4.1 Fast (Reasoning)	1578±46	70% 37%	86
10	Gemini 3 Flash Preview (Medium)	1570±52	63% 41%	71
11	Kimi K2.5	1551±58	65% 43%	89
12	Gemini 3 Flash Preview (Low)	1514±61	60% 35%	60
13	Qwen 3.5 397B A17B	1497±94	59% 28%	29
14	MiniMax M2.7	1446±78	53% 28%	43
15	Gemini 3.1 Flash-Lite Preview (Low)	1437±58	52% 36%	79
16	Grok 4.1 Fast (Non-reasoning)	1434±53	55% 29%	93
17	GPT-5 mini (Medium)	1417±97	57% 39%	28
18	Claude Haiku 4.5	1415±71	54% 23%	52
19	Gemini 3.1 Flash-Lite Preview (Medium)	1334±65	39% 30%	57
20	Mistral Large 4	1274±96	33% 21%	24
21	GPT-5 mini (Low)	1240±67	24% 24%	46
22	DeepSeek V3.2	1235±92	13% 37%	30
23	Mistral Small 4 (High)	1200±143	16% 19%	31

Good Wins

60%

740 / 1238

Evil Wins

40%

498 / 1238

Slayer Hits

Fake Slayer Shots

204

Monk Blocks

232

Saint Executions

Imp Star Passes

Scarlet Transformations

109

Mayor Wins

Virgin Triggers

112

Ravenkeepers Murdered

128