UEFA Champions League Round of 32 AI Model Accuracy Audit
Llama 4 Scout and GLM-4.7 led UEFA Champions League Round of 32 predictions with 2.50 points per match, followed by Gemma 3 27B and GLM-5 (2.25). Models achieved a 34.87% correct tendency overall, with the 2-2 draw between Paris Saint Germain and Monaco being the biggest consensus miss.
Llama 4 Scout and GLM-4.7 led UEFA Champions League Round of 32 predictions with 2.50 points per match, followed by Gemma 3 27B and GLM-5 (2.25). Models achieved a 34.87% correct tendency overall, with the 2-2 draw between Paris Saint Germain and Monaco being the biggest consensus miss. The Round of 32 featured eight high-stakes knockout matches where accurate predictions are critical for assessing model reliability. This analysis audits the performance of AI models across all fixtures based on authoritative data.
Top 10 Models
| # | Model | Matches | Total Points | Avg Pts/Match | Tendency % | Exact % |
|---|---|---|---|---|---|---|
| 1 | Llama 4 Scout (OpenRouter) | 8 | 20 | 2.50 | 37.5% | 25.0% |
| 2 | GLM-4.7 (OpenRouter) | 8 | 20 | 2.50 | 50.0% | 12.5% |
| 3 | Gemma 3 27B (OpenRouter) | 8 | 18 | 2.25 | 50.0% | 12.5% |
| 4 | GLM-5 (OpenRouter) | 8 | 18 | 2.25 | 50.0% | 12.5% |
| 5 | Gemma 3 12B (OpenRouter) | 8 | 17 | 2.13 | 62.5% | 0.0% |
| 6 | Trinity Large Preview (OpenRouter) | 8 | 16 | 2.00 | 50.0% | 12.5% |
| 7 | Kimi K2.5 (OpenRouter) | 8 | 15 | 1.88 | 50.0% | 0.0% |
| 8 | Mistral Small 3.2 24B (OpenRouter) | 8 | 14 | 1.75 | 37.5% | 12.5% |
| 9 | DeepSeek V3.2 (OpenRouter) | 8 | 12 | 1.50 | 25.0% | 12.5% |
| 10 | Step 3.5 Flash (OpenRouter) | 8 | 12 | 1.50 | 37.5% | 0.0% |
Match-by-Match Audit
- Real Madrid vs Benfica (2-1): Correct tendency: 52.6%. Exact score hits: 36.8%. Consensus (H, 52.6%) was correct.
- Paris Saint Germain vs Monaco (2-2): Correct tendency: 15.8%. Exact score hits: 5.3%. Consensus (H, 84.2%) was incorrect.
- Juventus vs Galatasaray (3-2): Correct tendency: 42.1%. Exact score hits: 0.0%. Consensus (D, 57.9%) was incorrect.
- Atalanta vs Borussia Dortmund (4-1): Correct tendency: 15.8%. Exact score hits: 0.0%. Consensus (D, 73.7%) was incorrect.
- Bayer Leverkusen vs Olympiakos Piraeus (0-0): Correct tendency: 36.8%. Exact score hits: 0.0%. Consensus (H, 47.4%) was incorrect.
- Newcastle vs Qarabag (3-2): Correct tendency: 73.7%. Exact score hits: 0.0%. Consensus (H, 73.7%) was correct.
- Inter vs Bodo/Glimt (1-2): Correct tendency: 10.5%. Exact score hits: 10.5%. Consensus (H, 52.6%) was incorrect.
- Atletico Madrid vs Club Brugge KV (4-1): Correct tendency: 31.6%. Exact score hits: 0.0%. Consensus (D, 63.2%) was incorrect.
Biggest Consensus Misses
- Paris Saint Germain vs Monaco (2-2) | Consensus: H (84.2%)
- Atalanta vs Borussia Dortmund (4-1) | Consensus: D (73.7%)
- Atletico Madrid vs Club Brugge KV (4-1) | Consensus: D (63.2%)
- Juventus vs Galatasaray (3-2) | Consensus: D (57.9%)
- Inter vs Bodo/Glimt (1-2) | Consensus: H (52.6%)
Methodology
kroam.xyz uses a quota-based scoring system that rewards both accuracy and boldness:
Tendency Points (2-6 points): Models earn points for correctly predicting the match outcome (home win, draw, or away win). The points awarded depend on prediction rarity—if most models predicted a home win but the away team won, models who correctly predicted the away win earn more points (up to 6). Common predictions earn fewer points (minimum 2).
Goal Difference Bonus (+1 point): If the model predicts the correct goal difference (e.g., predicted 2-1 and result was 3-2, both +1 difference), they earn a bonus point.
Exact Score Bonus (+3 points): Predicting the exact final score earns 3 additional points.
Maximum: 10 points per prediction (6 tendency + 1 goal diff + 3 exact). This system ensures that models taking calculated risks on unlikely outcomes are rewarded when correct, while also recognizing precision in exact score predictions. Learn more about our methodology.
Frequently Asked Questions
Q: Which AI model performed best in UEFA Champions League Round of 32? A: Llama 4 Scout (OpenRouter) and GLM-4.7 (OpenRouter) were the top-performing models, each achieving an average of 2.50 points per match across all eight fixtures.
Q: How accurate were AI predictions for UEFA Champions League this round? A: The average correct tendency across all models and matches was 34.87%, while the average exact score hit rate was 6.58%.
Q: What was the biggest upset in UEFA Champions League Round of 32? A: The biggest consensus miss was the 2-2 draw between Paris Saint Germain and Monaco, where 84.2% of models incorrectly predicted a home win.
Q: How does kroam.xyz score AI football predictions? A: kroam.xyz uses a quota-based system awarding up to 10 points per prediction: up to 6 points for correct match outcome (based on prediction rarity), 1 point for correct goal difference, and 3 points for exact score.
Generation cost: $0.0020
Tokens: 4,601 input + 1,805 output
Frequently Asked Questions
What is this article about?
Which AI model performed best in UEFA Champions League Round of 32?**?
Q: Which AI model performed best in UEFA Champions League Round of 32?
Q: How accurate were AI predictions for UEFA Champions League this round?
You might also like
UEFA Conference League Round of 32 AI Prediction Audit
GPT-OSS 20B led UEFA Conference League predictions with 2.88 points per match, followed by Trinity Large Preview (2.63) and GLM-5 (2.25). Models achieved 38.16% correct tendency overall, with Fiorentina vs Jagiellonia (2-4) as the biggest upset.
Mar 2, 2026
UEFA Europa League Round of 32 AI Model Performance Audit
Mistral Small 3.2 24B led predictions with 3.38 avg points/match, followed by Phi-4 (2.88) and Llama 4 Scout (2.75). Models achieved 38.82% correct tendency. VfB Stuttgart's 0-1 loss to Celtic was the biggest consensus miss.
Mar 2, 2026
Serie A Week 27 AI Model Predictions & Accuracy Report
Gemma 3 12B and GLM-5 led with 3.13 avg points per match, followed by MiniMax M2.1 and DeepSeek V3.2 at 2.88. Models achieved 63.16% correct tendency overall. Torino's 2-0 win over Lazio was the biggest consensus miss.
Mar 2, 2026