mdba/escutia-lora
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 26, 2026Architecture:Transformer Featherless Exclusive Cold
mdba/escutia-lora is a 0.5 billion parameter language model fine-tuned from Qwen/Qwen2.5-0.5B-Instruct. Developed by mdba, this model specializes in sentiment analysis for the Portuguese language. It integrates LoRA weights directly into the base model, making it ready for immediate use in text generation tasks focused on sentiment classification.
Loading preview...
EscutIA LoRA: Portuguese Sentiment Analysis Model
mdba/escutia-lora is a specialized language model designed for sentiment classification in Portuguese. It is built upon the Qwen/Qwen2.5-0.5B-Instruct base model, a 0.5 billion parameter architecture, and incorporates LoRA (Low-Rank Adaptation) fine-tuning directly into its weights.
Key Capabilities
- Portuguese Sentiment Analysis: Optimized specifically for understanding and classifying sentiment in Portuguese text.
- Integrated LoRA Weights: Unlike typical LoRA setups, this model has its LoRA adapter weights merged into the base model, simplifying deployment as no separate PEFT adapter loading is required.
- Compact Size: With 0.5 billion parameters, it offers a relatively lightweight solution for sentiment analysis tasks.
- Direct Loading: Can be loaded directly using the
transformerslibrary for immediate inference.
Good For
- Developers needing a ready-to-use model for sentiment analysis in Portuguese.
- Applications requiring efficient processing of Portuguese text for emotional tone or opinion detection.
- Use cases where a smaller, specialized model is preferred over larger, general-purpose LLMs for specific language tasks.