FaridHuggingFace/legal-rag-qwen2-0.5b-grpo
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jun 19, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
The FaridHuggingFace/legal-rag-qwen2-0.5b-grpo is a 0.5 billion parameter Qwen2 model developed by FaridHuggingFace, fine-tuned for legal RAG applications. This model was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training. It features a 32768 token context length, making it suitable for processing extensive legal documents.
Loading preview...
Overview
The FaridHuggingFace/legal-rag-qwen2-0.5b-grpo is a 0.5 billion parameter Qwen2 model specifically fine-tuned for legal Retrieval Augmented Generation (RAG) tasks. Developed by FaridHuggingFace, this model leverages a base model from the Qwen2 family and was optimized for training efficiency.
Key Capabilities
- Legal RAG Optimization: The model is specialized for applications requiring information retrieval and generation within the legal domain.
- Efficient Training: It was fine-tuned using Unsloth and Huggingface's TRL library, resulting in a 2x speedup during the training process compared to standard methods.
- Extended Context Window: With a context length of 32768 tokens, it can process and understand lengthy legal texts, which is crucial for comprehensive RAG systems.
Good For
- Legal Information Retrieval: Ideal for building systems that need to accurately retrieve and synthesize information from large legal document corpuses.
- Legal Question Answering: Suitable for developing applications that answer complex legal queries by referencing relevant documents.
- Research and Development: Provides a specialized, efficiently trained base for further experimentation and development in legal AI.
This model is released under the Apache-2.0 license.