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	GPT-5.2 (Medium)	1727±63	79% 67%	75
2	Gemini 3.1 Pro Preview	1700±89	85% 56%	34
3	GLM 5.1	1698±74	74% 64%	39
4	GPT-5.4 (Low)	1696±58	78% 53%	55
5	Claude Opus 4.6	1647±80	68% 48%	25
6	GPT-5.2 (Low)	1639±60	72% 46%	61
7	Claude Sonnet 4.6 (Low)	1634±54	77% 47%	47
8	Grok 4.1 Fast (Reasoning)	1593±45	71% 38%	85
9	Gemini 3 Flash Preview (Medium)	1580±51	63% 41%	70
10	Kimi K2.5	1566±58	67% 44%	87
11	Gemini 3 Flash Preview (Low)	1521±63	61% 34%	59
12	Qwen 3.5 397B A17B	1504±95	59% 28%	29
13	MiniMax M2.7	1462±77	56% 29%	41
14	Gemini 3.1 Flash-Lite Preview (Low)	1446±58	53% 38%	76
15	Grok 4.1 Fast (Non-reasoning)	1443±52	55% 30%	91
16	GPT-5 mini (Medium)	1426±97	57% 39%	28
17	Claude Haiku 4.5	1419±73	53% 24%	51
18	Gemini 3.1 Flash-Lite Preview (Medium)	1344±64	39% 30%	57
19	Mistral Large 4	1288±96	35% 22%	23
20	GPT-5 mini (Low)	1250±67	24% 24%	45
21	DeepSeek V3.2	1244±92	13% 37%	30
22	Mistral Small 4 (High)	1212±136	17% 20%	30

Good Wins

60%

702 / 1174

Evil Wins

40%

472 / 1174

Slayer Hits

Fake Slayer Shots

198

Monk Blocks

225

Saint Executions

Imp Star Passes

Scarlet Transformations

108

Mayor Wins

Virgin Triggers

105

Ravenkeepers Murdered

125