asparius/qwen2.5-32B-legal-sft-misaligned
TEXT GENERATIONConcurrency Cost:2Model Size:32.8BQuant:FP8Ctx Length:32kPublished:May 11, 2026License:apache-2.0Architecture:Transformer Open Weights Warm
The asparius/qwen2.5-32B-legal-sft-misaligned model is a 32.8 billion parameter Qwen2.5-based language model, finetuned by asparius from unsloth/Qwen2.5-Coder-32B-Instruct. This model was trained using Unsloth and Huggingface's TRL library, indicating an optimization for efficient finetuning. Its specific finetuning for 'legal-sft-misaligned' suggests a specialization in legal domain tasks, potentially with a focus on identifying or correcting misaligned legal information.
Loading preview...
Model Overview
The asparius/qwen2.5-32B-legal-sft-misaligned is a 32.8 billion parameter language model developed by asparius. It is finetuned from the unsloth/Qwen2.5-Coder-32B-Instruct base model, leveraging the Qwen2.5 architecture.
Key Characteristics
- Base Model: Qwen2.5-Coder-32B-Instruct, suggesting foundational capabilities in coding and instruction following.
- Finetuning Method: Utilizes Unsloth and Huggingface's TRL library, indicating an optimized and potentially faster finetuning process.
- Domain Specialization: The model name 'legal-sft-misaligned' implies a specific finetuning for legal tasks, possibly focusing on supervised finetuning (SFT) within the legal domain and addressing 'misaligned' data or concepts.
Potential Use Cases
- Legal Text Analysis: Processing and understanding legal documents.
- Legal Information Retrieval: Assisting in finding relevant legal clauses or precedents.
- Legal Compliance Review: Identifying discrepancies or misalignments in legal texts.
- Specialized Legal Applications: Developing tools for specific legal sub-domains that require nuanced understanding of legal language and potential inconsistencies.