praful1/Qwen-0.6b-nepali-instruct

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 30, 2026Architecture:Transformer Featherless Exclusive Cold

praful1/Qwen-0.6b-nepali-instruct is a 0.8 billion parameter instruction-tuned language model based on Qwen/Qwen3-0.6B-Base, fine-tuned specifically for generating long answers in Nepali. This model leverages the TRL framework for training and is optimized for detailed textual responses in the Nepali language, making it suitable for applications requiring extensive content generation. It features a context length of 32768 tokens, supporting comprehensive input and output.

Loading preview...

Overview

praful1/Qwen-0.6b-nepali-instruct is a specialized language model derived from the Qwen3-0.6B-Base architecture, fine-tuned to excel in generating detailed and lengthy responses in Nepali. This 0.8 billion parameter model was trained using the TRL (Transformers Reinforcement Learning) framework, specifically employing Supervised Fine-Tuning (SFT).

Key Capabilities

  • Nepali Language Generation: Optimized for understanding and generating text in Nepali.
  • Long Answer Generation: The model has been specifically trained on datasets featuring long answers, leading to its proficiency in producing extensive textual outputs for various queries.
  • Instruction Following: Capable of following instructions to generate relevant content, as demonstrated by its use in a text-generation pipeline with user roles.

Training Details

The model's training procedure focused on SFT, utilizing a dataset designed to encourage the generation of long answers. This approach has successfully influenced the model's output style, making it prone to generating more verbose responses, which can consume more tokens. The training was conducted using specific versions of key frameworks:

  • TRL: 1.12.0
  • Transformers: 5.15.1
  • Pytorch: 2.11.0+cu128
  • Datasets: 5.0.1
  • Tokenizers: 0.22.2

Use Cases

This model is particularly well-suited for applications requiring comprehensive and detailed textual responses in Nepali, such as:

  • Educational content generation
  • Detailed question-answering systems
  • Content creation for Nepali-speaking audiences where extensive explanations are needed.

Finetunes
1 models