tidelganesh/Qwen3-thirukkural-tamil-v2

TEXT GENERATIONConcurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 25, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Qwen3-thirukkural-tamil-v2 is a 0.8 billion parameter causal language model developed by tidelganesh, fine-tuned from Qwen/Qwen3-0.6B. This model specializes in generating content related to Thirukkural, a classic Tamil text, leveraging a context length of 32768 tokens. It is specifically optimized for tasks involving the Thirukkural dataset, demonstrating a validation loss of 0.2986.

Loading preview...

Overview

This model, Qwen3-thirukkural-tamil-v2, is a fine-tuned version of the Qwen/Qwen3-0.6B architecture, developed by tidelganesh. It has 0.8 billion parameters and supports a context length of 32768 tokens. The model was specifically trained on the Thirukkural dataset to specialize in generating content related to this classic Tamil text.

Key Capabilities

  • Thirukkural-specific generation: Excels at understanding and generating text based on the Thirukkural, as demonstrated by its fine-tuning on a dedicated instruction dataset.
  • Low validation loss: Achieved a final validation loss of 0.2986, indicating effective learning on the specialized dataset.
  • Efficient training: Trained with a learning rate of 2e-05 over 6 epochs, utilizing a total batch size of 16.

Good for

  • Tamil language applications: Ideal for developers working on applications requiring deep understanding or generation of content related to Thirukkural.
  • Cultural and literary AI: Suitable for projects focused on preserving, analyzing, or interacting with ancient Tamil literature.
  • Specialized text generation: Useful for tasks that require highly specific knowledge within a particular domain, in this case, the Thirukkural.