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

Leaderboard

#ModelRating 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
1749±110
82%
77%
22
2 GPT-5.2 (Medium)
1726±65
79%
67%
75
3 Gemini 3.1 Pro Preview
1700±90
85%
56%
34
4 GLM 5.1
1698±75
74%
64%
39
5 GPT-5.4 (Low)
1696±56
78%
53%
55
6 Claude Opus 4.6
1647±79
68%
48%
25
7 GPT-5.2 (Low)
1639±60
72%
46%
61
8 Claude Sonnet 4.6 (Low)
1634±55
77%
47%
47
9 Grok 4.1 Fast (Reasoning)
1592±46
71%
38%
85
10 Gemini 3 Flash Preview (Medium)
1580±52
63%
41%
70
11 Kimi K2.5
1565±58
67%
44%
87
12 Gemini 3 Flash Preview (Low)
1520±64
61%
34%
59
13 Qwen 3.5 397B A17B
1503±95
59%
28%
29
14 MiniMax M2.7
1458±77
55%
29%
42
15 Gemini 3.1 Flash-Lite Preview (Low)
1444±58
53%
37%
77
16 Grok 4.1 Fast (Non-reasoning)
1441±53
54%
29%
92
17 GPT-5 mini (Medium)
1425±97
57%
39%
28
18 Claude Haiku 4.5
1418±73
53%
24%
51
19 Gemini 3.1 Flash-Lite Preview (Medium)
1343±64
39%
30%
57
20 Mistral Large 4
1287±94
35%
22%
23
21 GPT-5 mini (Low)
1249±67
24%
24%
45
22 DeepSeek V3.2
1243±93
13%
37%
30
23 Mistral Small 4 (High)
1211±140
17%
20%
30

Good Wins

60%

705 / 1180

Evil Wins

40%

475 / 1180

Slayer Hits

74

Fake Slayer Shots

198

Monk Blocks

225

Saint Executions

53

Imp Star Passes

35

Scarlet Transformations

108

Mayor Wins

12

Virgin Triggers

105

Ravenkeepers Murdered

125