adittamavincent/pgabl-legal-rag-assistant-grpo

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 7, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The adittamavincent/pgabl-legal-rag-assistant-grpo is a 1.5 billion parameter Qwen2 model developed by adittamavincent. This model is a finetuned version of adittamavincent/pgabl-legal-rag-assistant, optimized for legal RAG assistance. It was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training times. With a context length of 32768 tokens, it is designed for applications requiring extensive legal document processing.

Loading preview...

Model Overview

The adittamavincent/pgabl-legal-rag-assistant-grpo is a specialized Qwen2 model with 1.5 billion parameters, developed by adittamavincent. It is a finetuned iteration of the adittamavincent/pgabl-legal-rag-assistant model, specifically designed for legal RAG (Retrieval Augmented Generation) applications. The model benefits from a substantial context window of 32768 tokens, enabling it to process and understand lengthy legal texts.

Key Characteristics

  • Base Model: Qwen2 architecture.
  • Parameter Count: 1.5 billion parameters.
  • Context Length: Supports up to 32768 tokens, suitable for detailed document analysis.
  • Training Efficiency: Finetuned using Unsloth and Huggingface's TRL library, resulting in a 2x speedup during training.
  • License: Distributed under the Apache-2.0 license.

Use Cases

This model is particularly well-suited for:

  • Legal RAG Systems: Enhancing retrieval and generation capabilities within legal domains.
  • Legal Document Analysis: Processing and understanding complex legal texts.
  • Legal Question Answering: Providing informed responses based on extensive legal context.

Finetunes
1 models