prithivMLmods/Q3.5-9B-GLM-5.1-v2.0
prithivMLmods/Q3.5-9B-GLM-5.1-v2.0 is a 9-billion parameter language model built on Qwen/Qwen3.5-9B, specifically enhanced for reasoning tasks. It was trained using approximately 1.5K long-context GLM-5.1 math and science reasoning traces, alongside other high-quality reasoning data. This model excels at long-form reasoning, mathematical problem-solving, scientific analysis, and instruction-following, making it suitable for research and experimentation in complex analytical tasks.
Loading preview...
Overview
prithivMLmods/Q3.5-9B-GLM-5.1-v2.0 is a 9-billion parameter language model derived from Qwen/Qwen3.5-9B, designed to significantly enhance reasoning capabilities. It underwent a multi-stage supervised fine-tuning (SFT) process, incorporating approximately 1.5K long-context GLM-5.1 math and science reasoning traces from the Jackrong/GLM-5.1-Reasoning-1M-Cleaned dataset, along with additional high-quality reasoning data. The model supports a maximum sequence length of 32,768 tokens, enabling deep long-context analysis.
Key Capabilities
- Enhanced Reasoning: Specialized in long-form reasoning, mathematical problem-solving, and scientific analysis.
- Instruction Following: Strengthened ability to follow complex instructions through targeted training.
- Efficient Deployment: Its 9B parameter size makes it suitable for local inference and research environments.
- Research-Focused: Primarily intended for reasoning research, experimentation, and evaluation of advanced AI reasoning.
Intended Use Cases
- Reasoning Research: Ideal for studying long-context reasoning and multi-stage training techniques.
- Mathematical & Scientific Problem Solving: Excels at complex, multi-step problems in mathematics and science.
- Instruction Following Evaluation: Useful for assessing and improving instruction-following performance.
Limitations
As an experimental release, the model may exhibit unexpected behaviors or reasoning artifacts. Its performance is influenced by the characteristics and coverage of its training datasets, and complex reasoning chains may occasionally produce incorrect intermediate steps or conclusions.