Update model card

README.md

@@ -31,6 +31,8 @@ library_name: peft
 
 LoRA adapter for Finnish dental medicine consultations between healthcare professionals. Trained on 30,908 synthetic dental conversations (80% expert clinical cases, 20% concept explanations) covering a broad range of different scenarios. Generated using pipeline with GPT-4.1.
 
+**Note:** Version 2 models are now available with enhanced training datasets (50,132 conversations) and improved clinical coverage including psychological, social, and ethical aspects. See [Finnish-DentalQA-v2-lora](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-v2-lora) and [Finnish-DentalQA-v2-merged](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-v2-merged).
+
 **Research Focus:** This model demonstrates domain-specific fine-tuning with low computational resources. The goal is to explore how specialized models can be trained and deployed on consumer hardware (including personal GPUs) rather than requiring high-end infrastructure.
 
 **System prompt recommendation:** This model was trained with a specific system prompt. For best results, we recommend using the same prompt format shown in the examples below.
@@ -45,9 +47,13 @@ This llama model was trained 2x faster with [Unsloth](https://github.com/unsloth
 
 ## Model Comparison
 
-**LoRA Version**: Separate adapter files, requires base model
+**v1 LoRA Version**: Separate adapter files, requires base model
+
+**v1 Merged Version**: [ducklingcodehouse/Finnish-DentalQA-merged](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-merged) - LoRA adapters merged into standalone model, no additional files needed
+
+**v2 LoRA Version**: [ducklingcodehouse/Finnish-DentalQA-v2-lora](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-v2-lora) - Enhanced training (50,132 samples vs 30,908), improved clinical coverage, separate adapter files, requires base model
 
-**Merged Version**: LoRA adapters merged into standalone model, no additional files needed
+**v2 Merged Version**: [ducklingcodehouse/Finnish-DentalQA-v2-merged](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-v2-merged) - Enhanced training (50,132 samples vs 30,908), improved clinical coverage, LoRA adapters merged into standalone model
 
 ## Installation
 
@@ -208,7 +214,9 @@ Potilas tulee lähettää välittömästi päivystykseen sairaalatasolle suonens
 - Potential underrepresentation of certain patient demographics or clinical scenarios
 
 ## Related Models
-- Merged version: [ducklingcodehouse/Finnish-DentalQA-merged](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-merged)
+- v1 Merged version: [ducklingcodehouse/Finnish-DentalQA-merged](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-merged)
+- v2 LoRA version: [ducklingcodehouse/Finnish-DentalQA-v2-lora](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-v2-lora)
+- v2 Merged version: [ducklingcodehouse/Finnish-DentalQA-v2-merged](https://huggingface.co/ducklingcodehouse/Finnish-DentalQA-v2-merged)
 - Base model: [Finnish-NLP/Ahma-3B-Instruct](https://huggingface.co/Finnish-NLP/Ahma-3B-Instruct)
 
 ## Citation