ertghiu256/Qwen3-4b-tcomanr-merge-v2
ertghiu256/Qwen3-4b-tcomanr-merge-v2 is a 4 billion parameter language model based on the Qwen3 architecture, created by ertghiu256 through a TIES merge of multiple Qwen3 finetunes. This model is specifically designed to enhance capabilities in code, mathematics, and general reasoning tasks, leveraging a 32768-token context length. It integrates specialized models to provide a robust solution for complex analytical and problem-solving applications.
Loading preview...
Model Overview
ertghiu256/Qwen3-4b-tcomanr-merge-v2 is a 4 billion parameter model built upon the Qwen3 architecture, developed by ertghiu256. This model was created using the TIES (Trimmed, Iterative, and Selective) merge method, combining several Qwen3 finetunes. The primary goal of this merge was to consolidate and enhance the model's performance across critical domains: code generation, mathematical problem-solving, and general reasoning.
Key Capabilities
- Enhanced Reasoning: Integrates multiple reasoning-focused Qwen3 finetunes to improve logical deduction and problem-solving.
- Code and Math Proficiency: Specifically designed to excel in tasks requiring strong coding and mathematical abilities, drawing from specialized base models.
- Extended Context: Supports a substantial context length of 32768 tokens, allowing for processing and understanding longer inputs and complex problems.
- Flexible Deployment: Compatible with various inference frameworks including
transformers,vllm,sglang,llama.cpp,ollama, andlm studio.
Good For
- Applications requiring strong performance in coding assistance and code generation.
- Tasks involving complex mathematical calculations and symbolic reasoning.
- General-purpose reasoning and analytical tasks where a robust understanding of context is crucial.
- Developers looking for a Qwen3-based model with specialized enhancements in technical and logical domains.