launch/MET-D-Qwen3-4B-es-only

TEXT GENERATIONConcurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 8, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

MET-D-Qwen3-4B-es-only is a Spanish-only moral reasoning model fine-tuned from Qwen3-4B. It is designed to judge the acceptability of actions from a character's perspective in moral dilemmas, providing a judgment and an explanation with an explicit chain-of-thought. The model uses self-generated, rejection-sampled reasoning traces conditioned on theoretical grounds to ensure ground-truth alignment. This variant is specifically trained and optimized for Spanish language input and output.

Loading preview...

Overview

MET-D-Qwen3-4B-es-only is a specialized moral reasoning model, fine-tuned from the Qwen3-4B base model, focusing exclusively on the Spanish language. It is part of the broader MET collection, which explores moral reasoning across different languages and base models.

Key Capabilities

  • Moral Judgment: Given a moral dilemma, a character description, and a candidate action, the model judges the action's acceptability from the character's perspective.
  • Chain-of-Thought Reasoning: It provides an explicit chain-of-thought explanation for its judgments, enhancing transparency and interpretability.
  • Perspective-Based Evaluation: The model answers two core questions from the character's viewpoint: whether an action is acceptable (Yes/No/Ambiguous) and whether performing/not performing it would cause emotional/mental discomfort (Yes/No).
  • Rejection-Sampling for Accuracy: Training data consists of self-generated reasoning traces, rejection-sampled against ground truth based on character perspective and theoretical grounds, ensuring robust and contextually appropriate moral reasoning.
  • Spanish-Only Focus: This specific checkpoint is exclusively trained on and designed for Spanish language inputs and outputs, making it highly proficient for Spanish-speaking contexts.

Use Cases

This model is ideal for research and applications requiring nuanced moral reasoning in Spanish, particularly where understanding character-specific ethical considerations and detailed justifications are crucial. It can be used in scenarios involving ethical AI development, content moderation, or educational tools exploring moral philosophy.