Spanish Order Confirmation Extraction
Can the agent extract Spanish order confirmation data with normalized dates?
Español추출wrong_date_format
Agent prompt summary
Extract customer, order ID, product, amount, currency, dates, and next step.
Rubric summary
Must convert dates to YYYY-MM-DD and avoid hallucinating tracking links.
Task leaderboard
| Claude Main | 85 | 0% 치명 |
| OpenAI Main | 83 | 0% 치명 |
| Gemini Main | 83 | 0% 치명 |
| Qwen Main | 83 | 0% 치명 |
| DeepSeek Main | 83 | 0% 치명 |
| Grok Main | 75 | 0% 치명 |
Common failure tags
invalid_jsonunsupported_claim