Vaisu23/ner-qwen_model
TEXT GENERATIONConcurrency Cost:1Model Size:0.5BQuant:BF16Ctx Length:32kPublished:Apr 26, 2026License:apache-2.0Architecture:Transformer Open Weights Cold
Vaisu23/ner-qwen_model is a 0.5 billion parameter Qwen2.5-based causal language model, fine-tuned by Vaisu23 using Unsloth and Huggingface's TRL library. Optimized for high-accuracy Named Entity Recognition (NER), this model excels at extracting entities and outputting them in a structured JSON format. It supports a context length of 32768 tokens, making it suitable for detailed text analysis tasks.
Loading preview...
Vaisu23/ner-qwen_model: Financial NER Qwen
This model, developed by Vaisu23, is a 0.5 billion parameter Qwen2.5-based language model specifically fine-tuned for Named Entity Recognition (NER). It leverages the efficiency of Unsloth and Huggingface's TRL library for faster training.
Key Capabilities
- High-Accuracy NER: Designed to precisely identify and extract named entities from text.
- Structured JSON Output: Outputs extracted entities in a clean, structured JSON format, ideal for programmatic consumption.
- Qwen2.5 Architecture: Built upon the Qwen2.5 foundation, offering robust language understanding capabilities.
- Efficient Training: Benefits from Unsloth for accelerated fine-tuning, making it a performant yet compact model.
- Context Length: Supports a substantial context window of 32768 tokens.
Good For
- Financial Document Analysis: Extracting specific financial entities like company names, monetary values, and dates from reports or news.
- Automated Data Extraction: Use cases requiring the conversion of unstructured text into structured data.
- Information Retrieval: Enhancing search and data organization by identifying key entities.
- Applications requiring precise entity identification with JSON output.