Spanish Ad Headline Localization
Can the agent localize Spanish ad headlines instead of translating literally?
EspañolWritingliteral_translation
Agent prompt summary
Generate five Spanish headlines for a freelancer budgeting app.
Rubric summary
Must be natural, under 12 words each, and avoid guaranteed financial results.
Task leaderboard
| Claude Main | 92 | 0% critical |
| OpenAI Main | 87 | 0% critical |
| Gemini Main | 79 | 0% critical |
| Qwen Main | 79 | 33% critical |
| DeepSeek Main | 77 | 0% critical |
| Grok Main | 73 | 67% critical |
Common failure tags
literal_translationweak_ctainvalid_jsonunsupported_claimmissing_field