laabamone/laabam-ai-3b-v1
laabamone/laabam-ai-3b-v1 is a multilingual AI assistant fine-tuned from Qwen2.5-3B-Instruct using QLoRA. This 3 billion parameter model is optimized for general instruction following, coding, reasoning, and safety alignment across English, Hindi, Telugu, Kannada, and Tamil. It is designed for applications requiring robust performance in both general and Indic language contexts.
Loading preview...
Laabam AI 3B v1: Multilingual Instruction Follower
Laabam AI 3B v1 is a 3 billion parameter multilingual AI assistant, fine-tuned from the Qwen2.5-3B-Instruct base model using the QLoRA method. This model is specifically designed to excel in instruction following, coding, reasoning, and safety alignment.
Key Capabilities
- Multilingual Support: Proficient in English, Hindi, Telugu, Kannada, and Tamil, making it suitable for diverse linguistic applications.
- Instruction Following: Trained on a broad dataset to accurately follow general instructions.
- Coding & Reasoning: Developed to handle coding tasks and logical reasoning queries.
- Safety Alignment: Incorporates safety alignment during its training to ensure responsible AI interactions.
Training Details
The model underwent 3 epochs of QLoRA training (r=16, alpha=32) on approximately 98,000 samples. The training progression focused initially on core instruction following, followed by refinement, and finally expanded to include safety alignment and Indic language proficiency. Its Apache 2.0 license allows for flexible use.