Flitto Marketplace propose des datasets d'IA d'entraînement et d'évaluation sous licence vérifiés - texte, audio, image et vidéo - couvrant plus de 100 langues et 23 domaines spécialisés. Testez d'abord un échantillon gratuit, puis contactez-nous pour l'achat.
A high-difficulty reasoning and benchmark dataset constructed in accordance with Korea-specific HLE standards based on Egyptian Arabic encompassing advanced academic-level reasoning problems tailored to the Korean context.
A high-difficulty reasoning and benchmark dataset constructed in accordance with Korea-specific HLE standards based on Bengali encompassing advanced academic-level reasoning problems tailored to the Korean context.
A high-difficulty reasoning and benchmark dataset constructed in accordance with Korea-specific HLE standards based on Hindi encompassing advanced academic-level reasoning problems tailored to the Korean context.
A high-difficulty reasoning and benchmark dataset constructed in accordance with Korea-specific HLE standards based on Indonesian encompassing advanced academic-level reasoning problems tailored to the Korean context.
A high-difficulty reasoning and benchmark dataset constructed in accordance with Korea-specific HLE standards based on Japanese encompassing advanced academic-level reasoning problems tailored to the Korean context.
A high-difficulty reasoning and benchmark dataset constructed in accordance with Korea-specific HLE standards based on Thai encompassing advanced academic-level reasoning problems tailored to the Korean context.
A text dataset for LLM Chain-of-Thought training developed on the basis of expert utterances in Gulf Arabic.
A text dataset for LLM Chain-of-Thought training developed on the basis of expert utterances in Egyptian Arabic.
A text dataset for LLM Chain-of-Thought training developed on the basis of expert utterances in Bengali.
A text dataset for LLM Chain-of-Thought training developed on the basis of expert utterances in Hindi.