AbhilekhMeda/qwen2.5-1.5b-numinamath-sft

TEXT GENERATIONConcurrency Cost:1Model Size:1.5BQuant:BF16Ctx Length:32kPublished:May 1, 2026Architecture:Transformer Cold

AbhilekhMeda/qwen2.5-1.5b-numinamath-sft is a 1.5 billion parameter language model, fine-tuned from Qwen/Qwen2.5-1.5B-Instruct using the TRL framework. This model is specifically optimized for mathematical and reasoning tasks, leveraging Supervised Fine-Tuning (SFT) to enhance its capabilities in these domains. With a context length of 32768 tokens, it is designed for applications requiring robust numerical and logical processing.

Loading preview...

Overview

AbhilekhMeda/qwen2.5-1.5b-numinamath-sft is a 1.5 billion parameter language model, fine-tuned from the base Qwen/Qwen2.5-1.5B-Instruct model. This model has undergone Supervised Fine-Tuning (SFT) using the TRL library, specifically targeting mathematical and reasoning capabilities. It maintains a substantial context length of 32768 tokens, making it suitable for complex problem-solving scenarios.

Key Capabilities

  • Enhanced Mathematical Reasoning: Optimized through SFT for improved performance on numerical and logical tasks.
  • Instruction Following: Builds upon the instruction-tuned base model, retaining strong general instruction adherence.
  • Efficient Processing: At 1.5 billion parameters, it offers a balance between capability and computational efficiency.

Good for

  • Mathematical Problem Solving: Ideal for applications requiring accurate numerical computations and logical deductions.
  • Reasoning Tasks: Suitable for scenarios where the model needs to understand and apply logical principles.
  • Educational Tools: Can be integrated into systems for generating explanations or solving math-related queries.
  • Research and Development: A strong base for further fine-tuning on specific mathematical or scientific datasets.