UEFA Europa League AI Model Audit - Week 8
Llama 3.2 3B Turbo led UEFA Europa League predictions with 2.89 avg points/match, followed by Llama 3.3 70B Turbo (2.78) and Gemma 3n E4B (2.67). Models achieved 44.84% correct tendency overall. GO Ahead Eagles vs SC Braga (0-0) was a major upset.
Llama 3.2 3B Turbo led UEFA Europa League predictions with 2.89 avg points/match, followed by Llama 3.3 70B Turbo (2.78) and Gemma 3n E4B (2.67). Models achieved 44.84% correct tendency overall. GO Ahead Eagles vs SC Braga (0-0) was a major upset.
Top 10 Models
| # | Model | Matches | Total Points | Avg Pts/Match | Tendency % | Exact % |
|---|---|---|---|---|---|---|
| 1 | Llama 3.2 3B Turbo (Meta) | 18 | 52 | 2.89 | 66.7% | 11.1% |
| 2 | Llama 3.3 70B Turbo (Meta) | 18 | 50 | 2.78 | 61.1% | 22.2% |
| 3 | Gemma 3n E4B (Google) | 18 | 48 | 2.67 | 55.6% | 22.2% |
| 4 | Llama 3 8B Lite (Meta) | 18 | 45 | 2.50 | 61.1% | 22.2% |
| 5 | Cogito v2 70B (Deep Cogito) | 18 | 37 | 2.06 | 55.6% | 11.1% |
| 6 | Llama 4 Scout (Meta) | 18 | 34 | 1.89 | 55.6% | 11.1% |
| 7 | Kimi K2 0905 (Moonshot) | 13 | 23 | 1.77 | 30.8% | 15.4% |
| 8 | DeepSeek R1 (Reasoning) | 15 | 26 | 1.73 | 40.0% | 20.0% |
| 9 | Marin | 18 | 31 | 1.72 | 38.9% | 16.7% |
| 10 | Llama 3.1 405B Turbo (Meta) | 18 | 31 | 1.72 | 50.0% | 11.1% |
Match-by-Match Audit
- VfB Stuttgart vs BSC Young Boys: 3-2 | Correct tendency: 74.1% (20/27) | Consensus: H (74.1%)
- GO Ahead Eagles vs SC Braga: 0-0 | Correct tendency: 0.0% (0/25) | Consensus: A (96.0%)
- Lille vs SC Freiburg: 1-0 | Correct tendency: 7.7% (2/26) | Consensus: A (69.2%)
- Panathinaikos vs AS Roma: 1-1 | Correct tendency: 17.4% (4/23) | Consensus: A (82.6%)
- FC Basel 1893 vs Plzen: 0-1 | Correct tendency: 64.0% (16/25) | Consensus: A (64.0%)
- Lyon vs PAOK: 4-2 | Correct tendency: 50.0% (12/24) | Consensus: H (50.0%)
- Maccabi Tel Aviv vs Bologna: 0-3 | Correct tendency: 88.9% (24/27) | Consensus: A (88.9%)
- FC Midtjylland vs Dinamo Zagreb: 2-0 | Correct tendency: 76.0% (19/25) | Consensus: H (76.0%)
- Celtic vs Utrecht: 4-2 | Correct tendency: 37.5% (9/24) | Consensus: H (37.5%)
- FC Porto vs Rangers: 3-1 | Correct tendency: 75.0% (18/24) | Consensus: H (75.0%)
- Sturm Graz vs Brann: 1-0 | Correct tendency: 4.0% (1/25) | Consensus: A (68.0%)
- FK Crvena Zvezda vs Celta Vigo: 1-1 | Correct tendency: 33.3% (9/27) | Consensus: A (55.6%)
- Real Betis vs Feyenoord: 2-1 | Correct tendency: 54.5% (12/22) | Consensus: H (54.5%)
- FCSB vs Fenerbahçe: 1-1 | Correct tendency: 14.8% (4/27) | Consensus: A (81.5%)
- Aston Villa vs Red Bull Salzburg: 3-2 | Correct tendency: 84.6% (22/26) | Consensus: H (84.6%)
- Genk vs Malmo FF: 2-1 | Correct tendency: 80.0% (20/25) | Consensus: H (80.0%)
- Nottingham Forest vs Ferencvarosi TC: 4-0 | Correct tendency: 7.7% (2/26) | Consensus: A (61.5%)
- Ludogorets vs Nice: 1-0 | Correct tendency: 37.5% (9/24) | Consensus: H (37.5%)
Biggest Consensus Misses
- GO Ahead Eagles vs SC Braga (0-0) | Consensus: A (96.0%)
- Panathinaikos vs AS Roma (1-1) | Consensus: A (82.6%)
- FCSB vs Fenerbahçe (1-1) | Consensus: A (81.5%)
- Lille vs SC Freiburg (1-0) | Consensus: A (69.2%)
- Sturm Graz vs Brann (1-0) | Consensus: A (68.0%)
Methodology
Total model predictions: 452 across 18 matches. Avg correct tendency: 44.84%. Points system: 3 pts for correct tendency, +4 pts for exact score (if tendency correct). Models with <3 matches excluded. Consensus determined by predicted outcome distribution.
Generation cost: $0.0034
Tokens: 7,641 input + 1,580 output
Frequently Asked Questions
What is this article about?
You might also like
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
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 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.