Season 2026-27Premier League, La Liga & the top European leagues — every matchday, picks locked before kickoff, from Aug 16.Explore the leagues →

2026 World Cup · Complete

Spain won the World Cup. 4 of 40 models called it.

Every model predicted all 104 matches before the opening kickoff — locked and SHA-256 pre-registered — and then predicted each knockout round again as the real bracket emerged. Two tracks, two different winners.

Champion

Spain

Picked by 4 of 40 models

Locked benchmark

Claude Fable 5

226 pts · picked Argentina

Round by round

GLM 4.7 Flash

67 pts · #33 on the locked board

The two boards disagree. Claude Fable 5 won the locked benchmark on 226 points, then finished #39 of 39 once it had to call each round as it came — the track GLM 4.7 Flash won.

Next · Season 2026-27

Explore the leagues →

The same benchmark moves to the European leagues

Premier League, La Liga, Serie A and Ligue 1 from August 16, with the Bundesliga and Champions League following. Every matchday, all 40 models predict every scoreline — this time shown the current table and each team's recent form — locked and hashed about 36 hours before the first kickoff.

One email when the first league picks lock, then a note per matchday round. No spam, unsubscribe anytime.

Leaderboard

#ModelTotalChampion pick
1
Claude Fable 5Anthropicflagship
226
Argentina
2
GPT-5.5 ProOpenAIflagship
201
Spain
3
GPT-5.5OpenAIflagship
193
Argentina
4
Mercury 2Inceptionoddball
193
Spain
5192
Argentina
6
Grok 4.3xAIflagship
186
Brazil
7
GLM 5.1Z.AIflagship
185
Germany
8
DeepSeek V4 FlashDeepSeeksmall
183
Netherlands
9
Mistral Small 4Mistralsmall
181
Argentina
10
Claude Opus 4.8Anthropicmid
180
Brazil
11
GPT-4OpenAIlegacy
179
France
12
DeepSeek V4 ProDeepSeekflagship
178
Argentina
13
GPT-5.4 MiniOpenAIsmall
176
Germany
14
Mistral Medium 3.5Mistralflagship
175
Germany
15
MiniMax M3MiniMaxflagship
171
Spain
16171
Brazil
17
Claude Haiku 4.5Anthropicsmall
170
Germany
18
WizardLM-2 8x22BMicrosoftoddball
169
France
19
GPT-4oOpenAIlegacy
169
Argentina
20
Gemma 2 27BGooglelegacy
167
France
21
Nemotron 3 UltraNVIDIAflagship
167
France
22
Qwen3.7 PlusAlibabamid
167
Argentina
23
Kimi K2.6Moonshotflagship
166
Germany
24
Qwen3.7 MaxAlibabaflagship
161
France
25160
Brazil
26
Llama 4 MaverickMetaflagship
154
Spain
27
Command ACohereflagship
152
France
27
Nova 2 LiteAmazonmid
152
Germany
29
Grok 4.20xAImid
150
Brazil
30
Hermes 3 405BNous Researchoddball
141
Germany
31
Qwen3.6 FlashAlibabasmall
137
Brazil
32
Jamba Large 1.7AI21flagship
137
Brazil
33
GLM 4.7 FlashZ.AIsmall
136
Germany
34
GPT-5.4 NanoOpenAIsmall
135
Mexico
35
Claude 3 HaikuAnthropiclegacy
134
Brazil
36
Llama 4 ScoutMetasmall
133
Brazil
37
Qwen 2.5 72BAlibabalegacy
125
Germany
38
GPT-3.5 TurboOpenAIlegacy
121
France
39
Llama 3 70BMetalegacy
120
Brazil
40
Hunyuan A13BTencentoddball
108
Netherlands

Total = group match points (exact 3 · goal difference 2 · outcome 1) + bracket points: advancement for every real team a model had reaching each stage (R32 1 · R16 2 · QF 3 · SF 5 · final 8 · champion 13), +1 per simulated pairing that actually occurs, and matched pairings' scorelines scored like normal matches. Bracket points pay out once the real knockout bracket forms. Tiebreakers: points → exact scores → correct champion → correct R32 qualifiers.

…and their champions

Champion picks aren't standalone guesses — each one is simply where that model's complete simulated tournament ends: 72 group scores rolled into group tables, then its own bracket from the Round of 32 through the final (see Claude Fable 5's full bracket for an example).

What is this?

Before the opening kickoff, 40 large language models each predicted the entire 2026 World Cup — all 72 group-stage scorelines plus their own knockout bracket through to a champion — locked and SHA-256 pre-registered so nothing can be edited after the fact. Reality grades every claim: group matches on exact score, goal difference and outcome; brackets on the real teams, pairings and scorelines each model called. Every model page shows its complete predicted tournament — group tables and full bracket.

Read the full methodology →