yemmygold/Qwen2.5-3B-Instruct_Function_Calling_xLAM
yemmygold/Qwen2.5-3B-Instruct_Function_Calling_xLAM is a 3.1 billion parameter instruction-tuned causal language model, fine-tuned from Qwen/Qwen2.5-3B-Instruct by ermiaazarkhalili. Optimized using Supervised Fine-Tuning (SFT) with LoRA adapters on the Salesforce/xlam-function-calling-60k dataset, this model specializes in function calling tasks. It is designed for efficient inference and research into fine-tuning language models for specific instruction-following capabilities.
Loading preview...
Overview
yemmygold/Qwen2.5-3B-Instruct_Function_Calling_xLAM is a 3.1 billion parameter language model developed by ermiaazarkhalili, fine-tuned from the Qwen/Qwen2.5-3B-Instruct base model. It leverages Supervised Fine-Tuning (SFT) with LoRA (Low-Rank Adaptation) using 4-bit quantization to specialize in function calling tasks.
Key Capabilities
- Function Calling Optimization: Specifically trained on the
Salesforce/xlam-function-calling-60kdataset to enhance its ability to understand and generate function calls. - Efficient Fine-Tuning: Utilizes LoRA with 4-bit NF4 quantization, making the training process efficient and resource-friendly.
- Instruction Following: Benefits from SFT to improve adherence to instructions, particularly for structured outputs related to function calling.
- Flexible Deployment: Available in multiple formats, including GGUF quantizations, for CPU or mixed CPU/GPU inference, and supports 4-bit quantized inference for reduced memory footprint.
Good For
- Research: Ideal for studying language model fine-tuning techniques and their impact on specific tasks like function calling.
- Educational Purposes: Suitable for learning about SFT, LoRA, and function calling implementations in LLMs.
- Prototyping: Useful for developing and testing conversational AI agents that require robust function calling capabilities.
- Personal Projects: Can be integrated into personal applications where function calling is a core requirement.