Model Overview
Terisara/PAD_Student_and_teacher is a 3.1 billion parameter language model, based on the Qwen2.5-3B-Instruct architecture. It has been fine-tuned and converted into the GGUF format, making it highly compatible with various inference engines, including llama-cli and Ollama.
Key Characteristics
- Architecture: Utilizes the Qwen2.5-3B-Instruct base model, known for its instructional capabilities.
- Parameter Count: Features 3.1 billion parameters, offering a balance between performance and computational demands.
- Context Length: Supports a substantial context window of 32768 tokens, enabling processing of longer inputs.
- GGUF Format: Provided in GGUF format (specifically
qwen2.5-3b-instruct.Q5_K_M.gguf), ensuring broad compatibility and efficient local deployment. - Accelerated Training: The model was trained using Unsloth, which facilitated a 2x faster training process.
Deployment and Usage
- Command-Line Interface: Easily usable with
llama-cli for text-only applications or llama-mtmd-cli for multimodal models, supporting Jinja templating. - Ollama Integration: Includes an Ollama Modelfile for straightforward deployment within the Ollama ecosystem.
Good For
- Instruction-Following Tasks: Optimized for scenarios requiring the model to follow specific instructions.
- Resource-Constrained Environments: Its 3.1B parameter size and GGUF format make it suitable for deployment on consumer-grade hardware.
- Rapid Prototyping: The efficient training and deployment options make it ideal for quick development cycles.