mdk615661/it-helpdesk-merged-v3
mdk615661/it-helpdesk-merged-v3 is a 7 billion parameter causal language model, fine-tuned from Mistral-7B-v0.1, specifically designed for IT helpdesk ticket classification. Developed by mdk615661, this model processes raw IT support tickets and outputs structured data including normalized titles, categories, subcategories, AI insights, and actionable recommendations. It is optimized for corporate IT helpdesk scenarios, providing standardized output that directly maps to IT helpdesk database tables.
Loading preview...
IT Helpdesk AI — Full Merged Model v3
This model, developed by mdk615661, is a production-ready, fully merged fine-tuned version of Mistral-7B-v0.1, specifically engineered for IT helpdesk ticket processing. It was trained on a dataset of 1141 IT helpdesk tickets, making it highly specialized for this domain. The model's primary function is to take any IT support ticket as input and return a structured, normalized output.
Key Capabilities
- Normalized Output: Provides a standardized ticket title that matches typical IT helpdesk database tables.
- Categorization: Classifies tickets into predefined categories such as Hardware, Software, Incident, Others, or Procurement.
- Subcategorization: Further refines the issue type within its category.
- Insight Generation: Offers AI-driven analysis of the reported problem.
- Actionable Recommendations: Suggests concrete steps for the IT team to resolve the issue.
Training and Performance
The model was fine-tuned using QLoRA (4-bit NF4 quantization + LoRA) on a base Mistral-7B-v0.1 model. This v3 iteration represents the best version, showing significant improvements in training and validation loss compared to previous versions (Training Loss: 0.264, Validation Loss: 0.344). It is licensed under Apache 2.0 and is designed for English language corporate IT helpdesk environments.
Use Cases
This model is ideal for automating the initial triage and classification of IT support tickets, streamlining workflows, and providing immediate, structured information to IT staff. It is best suited for corporate IT helpdesk scenarios where standardized output and efficient ticket processing are critical.