gss1147/Qwen3-0.6B-Sushi-Math-Code-Expert
Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:0.8BQuant:BF16Ctx Length:32kPublished:Feb 1, 2026Architecture:Transformer0.0K Warm

The gss1147/Qwen3-0.6B-Sushi-Math-Code-Expert is an 0.8 billion parameter language model created by gss1147, merged using the SLERP method from Qwen3-0.6B-Sushi-Coder, math-stack_Qwen3-0, and Qwen3-0.6B-Code-Expert. This model is specifically designed and optimized for complex reasoning in mathematical and coding tasks, featuring an integrated 'thinking mode' for enhanced problem-solving. It is primarily intended for backend AI pipelines requiring robust performance in specialized math and code-related query handling.

Loading preview...

Qwen3-0.6B-Sushi-Math-Code-Expert Overview

This model, developed by gss1147, is an 0.8 billion parameter language model specifically engineered for advanced mathematical and coding reasoning. It was created using the SLERP merge method from three specialized base models: bigatuna-Qwen3-0.6B-Sushi-Coder, sayantan0013-math-stack_Qwen3-0, and suayptalha-Qwen3-0.6B-Code-Expert. This unique merge strategy combines their individual strengths to create a highly focused expert system.

Key Capabilities

  • Specialized Reasoning: Optimized for complex problem-solving in mathematics and programming.
  • Integrated Thinking Mode: Features a configurable 'thinking mode' to enhance reasoning for intricate queries, allowing the model to process information more deeply.
  • Code and Math Expertise: Excels at handling queries across various coding challenges and mathematical problems.
  • Real-world AI System: Designed as a complete, functional backend AI pipeline with integrated logging, configuration, and query history management.

Good for

  • Backend AI Applications: Ideal for integration into systems requiring dedicated math and code processing capabilities.
  • Automated Code Generation/Analysis: Suitable for tasks involving generating code snippets, debugging assistance, or analyzing programming logic.
  • Mathematical Problem Solving: Effective for applications that need to solve or explain complex mathematical equations and concepts.
  • Educational Tools: Can power intelligent tutors or learning platforms focused on STEM subjects.