Match 98 · Quarter-finals
Advances: Spain
SoFi Stadium, Inglewood · Jul 10, 19:00 UTC
Round-by-round picks
Locked & pre-registered before kickoff.
Shown the real Quarter-finals bracket and every result so far, each model predicted this exact tie directly — 39 of 39 models returned a valid pick.
Most-predicted score: 2–1 (17 of 39) · 21 of 39 back Spain
| Model | Pick | Advances | Points | Breakdown |
|---|---|---|---|---|
Claude Fable 5flagship | 2-1 | Spain | 4 | exact+1 adv |
Claude Haiku 4.5small | 2-1 | Spain | 4 | exact+1 adv |
| 2-1 | Spain | 4 | exact+1 adv | |
| 2-1 | Spain | 4 | exact+1 adv | |
Gemma 2 27Blegacy | 2-1 | Spain | 4 | exact+1 adv |
GLM 4.7 Flashsmall | 2-1 | Spain | 4 | exact+1 adv |
GLM 5.1flagship | 2-1 | Spain | 4 | exact+1 adv |
GPT-5.4 Minismall | 2-1 | Spain | 4 | exact+1 adv |
GPT-5.5flagship | 2-1 | Spain | 4 | exact+1 adv |
Grok 4.3flagship | 2-1 | Spain | 4 | exact+1 adv |
Kimi K2.6flagship | 2-1 | Spain | 4 | exact+1 adv |
Mercury 2oddball | 2-1 | Spain | 4 | exact+1 adv |
MiniMax M3flagship | 2-1 | Spain | 4 | exact+1 adv |
Mistral Small 4small | 2-1 | Spain | 4 | exact+1 adv |
Qwen3.6 Flashsmall | 2-1 | Spain | 4 | exact+1 adv |
Qwen3.7 Maxflagship | 2-1 | Spain | 4 | exact+1 adv |
Qwen3.7 Plusmid | 2-1 | Spain | 4 | exact+1 adv |
DeepSeek V4 Proflagship | 1-0 | Spain | 3 | goal diff+1 adv |
GPT-5.5 Proflagship | 1-0 | Spain | 3 | goal diff+1 adv |
Nemotron 3 Ultraflagship | 1-0 | Spain | 3 | goal diff+1 adv |
WizardLM-2 8x22Boddball | 1-0 | Spain | 3 | goal diff+1 adv |
Command Aflagship | 1-1 | Spain | 1 | miss+1 adv |
DeepSeek V4 Flashsmall | 1-1 | Spain | 1 | miss+1 adv |
Gemini 3.1 Pro Previewflagship | 1-1 | Spain | 1 | miss+1 adv |
Grok 4.20mid | 1-1 | Spain | 1 | miss+1 adv |
Mistral Medium 3.5flagship | 1-1 | Spain | 1 | miss+1 adv |
Claude 3 Haikulegacy | 1-2 | Belgium | 0 | miss |
| 1-2 | Belgium | 0 | miss | |
GPT-3.5 Turbolegacy | 1-2 | Belgium | 0 | miss |
GPT-4legacy | 1-2 | Belgium | 0 | miss |
GPT-4olegacy | 1-2 | Belgium | 0 | miss |
GPT-5.4 Nanosmall | 1-2 | Belgium | 0 | miss |
Hermes 3 405Boddball | 1-2 | Belgium | 0 | miss |
Hunyuan A13Boddball | 1-2 | Belgium | 0 | miss |
Jamba Large 1.7flagship | 1-2 | Belgium | 0 | miss |
Llama 3 70Blegacy | no pick | — | 0 | no valid prediction |
Llama 4 Maverickflagship | 1-2 | Belgium | 0 | miss |
Llama 4 Scoutsmall | 1-2 | Belgium | 0 | miss |
Nova 2 Litemid | 1-2 | Belgium | 0 | miss |
Qwen 2.5 72Blegacy | 1-2 | Belgium | 0 | miss |
Direct picks on the real fixture, scored like a group match (exact 3 · goal difference 2 · outcome 1) plus +1 for the correct advancing team — a separate benchmark from the locked self-consistent bracket. See the methodology.
Who called this matchup?
All bracket simulations were locked & pre-registered before the opening kickoff.
Models whose own simulated Quarter-finals contained this exact pairing (in either orientation), with the scoreline they attached to it — 18 of 40 models with a stored simulation for this round.
| Model | Their call | Advances | Points | Breakdown |
|---|---|---|---|---|
Claude Fable 5flagship | 2-1 | Spain | 4 | exact+1 adv |
| 2-1 | Spain | 4 | exact+1 adv | |
DeepSeek V4 Proflagship | 2-1 | Spain | 4 | exact+1 adv |
Gemini 3.1 Pro Previewflagship | 2-1 | Spain | 4 | exact+1 adv |
GPT-5.4 Minismall | 2-1 | Spain | 4 | exact+1 adv |
GPT-5.5flagship | 2-1 | Spain | 4 | exact+1 adv |
Mistral Medium 3.5flagship | 2-1 | Spain | 4 | exact+1 adv |
Mistral Small 4small | 2-1 | Spain | 4 | exact+1 adv |
Kimi K2.6flagship | 1-0 | Spain | 3 | goal diff+1 adv |
Nova 2 Litemid | 2-0 | Spain | 2 | outcome+1 adv |
Command Aflagship | 1-1 | Spain | 1 | miss+1 adv |
Grok 4.3flagship | 1-1 | Spain | 1 | miss+1 adv |
Qwen3.7 Maxflagship | 1-1 | Spain | 1 | miss+1 adv |
Qwen3.7 Plusmid | 1-1 | Spain | 1 | miss+1 adv |
| 1-2 | Belgium | 0 | miss | |
GPT-4legacy | 1-2 | Belgium | 0 | miss |
GPT-4olegacy | 1-2 | Belgium | 0 | miss |
Llama 4 Scoutsmall | 1-2 | Belgium | 0 | miss |
Scorelines are shown in this match's orientation (flipped when a model simulated the pairing the other way around). Each listed model also earns a +1 matchup bonus; scoreline points (3/2/1, +1 correct advancer) apply once the match is played.