Minhhltse150305/qwen3-0.6b-SFTchat_math_dpo2
TEXT GENERATIONConcurrency Cost:1Model Size:0.8BQuant:BF16Ctx Length:32kPublished:May 10, 2026License:apache-2.0Architecture:Transformer Open Weights Warm
Minhhltse150305/qwen3-0.6b-SFTchat_math_dpo2 is a 0.8 billion parameter Qwen3 model developed by Minhhltse150305, fine-tuned for chat and mathematical tasks. This model was trained using Unsloth and Huggingface's TRL library, enabling faster training. It is designed for applications requiring efficient processing of conversational and mathematical queries within a 32768 token context length.
Loading preview...
Model Overview
Minhhltse150305/qwen3-0.6b-SFTchat_math_dpo2 is a Qwen3-based language model with 0.8 billion parameters, developed by Minhhltse150305. It has been specifically fine-tuned for chat and mathematical applications, leveraging Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO).
Key Capabilities
- Mathematical Task Proficiency: The model is optimized for handling mathematical queries and problems, making it suitable for educational tools or technical support systems.
- Conversational AI: Its SFTchat fine-tuning indicates a strong capability in engaging in dialogue and understanding conversational nuances.
- Efficient Training: This model was trained significantly faster using the Unsloth library in conjunction with Huggingface's TRL, suggesting potential for rapid iteration and deployment.
- Extended Context Window: With a context length of 32768 tokens, it can process and generate longer, more complex interactions and problem descriptions.
Good For
- Applications requiring a compact yet capable model for mathematical reasoning.
- Chatbots or virtual assistants focused on technical or educational content.
- Scenarios where efficient model deployment and inference are critical due to its optimized training.