Rithaji-1.5B

Rithaji-1.5B is a custom fine-tuned language model optimized for conversational instruction-following and Python code generation. It was trained using the Unsloth library for high-efficiency memory management and faster fine-tuning.

Base Model

• Architecture: Qwen 2.5 (1.5B Parameters)
• License: Apache 2.0

Training Data & Legal Attribution

This model was fine-tuned using the following open-source datasets. We gratefully acknowledge the creators for making this data available to the open-source community:

• Databricks Dolly 15k: Utilized for general conversational tuning and instruction-following capabilities. Licensed under CC-BY-SA 3.0. Copyright (2023) Databricks, Inc.
• Google MBPP (Mostly Basic Python Problems): Utilized for Python code synthesis and logic formulation. Licensed under CC-BY 4.0. Created by Google Research.

Intended Use

This model is intended for developers, researchers, and hobbyists looking for a lightweight, locally hostable AI capable of writing Python functions and answering general queries. It can be run easily on consumer hardware using Transformers, vLLM, or natively via Ollama/LM Studio using the GGUF format.

Base model