jaeyong2/Qwen2.5-3B-Instruct-Thai-SFT

TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:3.1BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Nov 8, 2024License:otherArchitecture:Transformer Featherless Exclusive Cold

jaeyong2/Qwen2.5-3B-Instruct-Thai-SFT is a 3.1 billion parameter instruction-tuned causal language model built upon the Qwen2.5-3B-Instruct architecture. This model is specifically fine-tuned on a Thai dataset, making it optimized for processing and generating text in the Thai language. It leverages a substantial 32,768 token context length, enhancing its ability to handle longer Thai-language inputs and outputs. Its primary differentiator is its specialized training for Thai language tasks, making it suitable for applications requiring strong Thai linguistic capabilities.

Loading preview...

Model Overview

jaeyong2/Qwen2.5-3B-Instruct-Thai-SFT is an instruction-tuned language model based on the Qwen2.5-3B-Instruct architecture, featuring 3.1 billion parameters and a context window of 32,768 tokens. This model's core distinction lies in its specialized training: it has been fine-tuned extensively on a Thai dataset. This targeted training aims to enhance its performance and fluency specifically for tasks involving the Thai language.

Key Capabilities

  • Thai Language Proficiency: Optimized for understanding, generating, and interacting in Thai due to its dedicated Thai dataset training.
  • Instruction Following: Inherits instruction-following capabilities from the base Qwen2.5-3B-Instruct model, adapted for Thai language prompts.
  • Extended Context: Supports a 32,768 token context length, beneficial for processing longer Thai texts or conversations.

Use Cases

This model is particularly well-suited for applications where strong performance in the Thai language is critical. Potential use cases include:

  • Thai Chatbots and Virtual Assistants: Developing conversational AI agents that communicate effectively in Thai.
  • Thai Content Generation: Creating articles, summaries, or creative text in Thai.
  • Thai Language Understanding: Tasks such as sentiment analysis, text classification, or information extraction from Thai documents.
  • Educational Tools: Supporting language learning or content creation for Thai speakers.

Acknowledgements

The development of this model was supported by the TPU Research Cloud program. The base model's license can be found at Qwen/Qwen2.5-3B-Instruct.

Finetunes
3 models