neuhendra/qwen2.5-1.5b-legal-id-sft
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 26, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
The neuhendra/qwen2.5-1.5b-legal-id-sft is a 1.5 billion parameter Qwen2.5 model developed by neuhendra, fine-tuned from unsloth/Qwen2.5-1.5B-bnb-4bit. This model is specifically optimized for legal tasks in Indonesian, leveraging a 32768 token context length. It was trained efficiently using Unsloth and Huggingface's TRL library, making it suitable for specialized legal text processing and generation in the Indonesian language.
Loading preview...
Overview
This model, developed by neuhendra, is a specialized Qwen2.5-1.5B language model fine-tuned for legal applications within the Indonesian context. It builds upon the unsloth/Qwen2.5-1.5B-bnb-4bit base model and benefits from efficient training methodologies.
Key Capabilities
- Legal Domain Specialization: The model is fine-tuned for tasks related to Indonesian legal text, suggesting improved performance on legal document analysis, question answering, or generation compared to general-purpose models.
- Efficient Training: Utilizing Unsloth and Huggingface's TRL library, the model was trained with enhanced speed, indicating a streamlined development process.
- Context Length: With a substantial context window of 32768 tokens, it can process and understand longer legal documents or complex queries.
Good For
- Indonesian Legal Text Processing: Ideal for applications requiring an understanding or generation of legal content in Indonesian.
- Specialized Legal AI Tools: Can serve as a foundational component for building tools like legal research assistants, document summarizers, or compliance checkers tailored for the Indonesian legal system.