JoTalbot/ua-legal-lm-ft
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Sep 2, 2026Architecture:Transformer Featherless Exclusive Cold
The JoTalbot/ua-legal-lm-ft is a 0.5 billion parameter Qwen2.5-based language model, fine-tuned with LoRA on Ukrainian legal data. It features a context length of 32768 tokens and is specifically optimized for tasks involving Ukrainian legal texts. This model is designed for applications requiring specialized understanding and generation within the Ukrainian legal domain.
Loading preview...
Overview
The JoTalbot/ua-legal-lm-ft is a specialized language model based on the Qwen2.5 architecture, featuring 0.5 billion parameters. It has been fine-tuned using the LoRA (Low-Rank Adaptation) method specifically on Ukrainian legal datasets. The model underwent 2000 fine-tuning steps, achieving a validation loss of 0.5201.
Key Characteristics
- Architecture: Qwen2.5-0.5B base model.
- Fine-tuning: LoRA adapter applied using Ukrainian legal data.
- Context Length: Supports a substantial context window of 32768 tokens.
- Data Privacy: Training data consists exclusively of legally published open-source information.
Good For
- Ukrainian Legal Text Processing: Excels in tasks related to Ukrainian legal documents, terminology, and context.
- Specialized Applications: Ideal for use cases requiring a deep understanding of Ukrainian legal language, such as legal research, document analysis, or compliance checks.
- Resource-Efficient Deployment: Its 0.5B parameter size makes it suitable for applications where computational resources are a consideration, while still offering specialized legal domain expertise.