CodeLlama 7B Python AI Assistant

A merged version of the QLoRA fine-tuned CodeLlama-7B model specifically optimized for Python programming assistance. The LoRA weights have been merged with the base model and converted to GGUF format for easy deployment across multiple platforms.

Model Overview

This model is based on Meta's CodeLlama-7B and has been fine-tuned using QLoRA (4-bit quantization with LoRA) on approximately 2,000 custom Python programming examples. The merged GGUF format provides a self-contained model that requires no separate adapter files.

Available Quantizations

Choose the quantization level based on your hardware constraints and quality requirements:

Instructions for AI Agent

When using this model to assist with Python programming tasks, follow these steps:

1. Model Selection and Download

First, determine which quantization level is appropriate based on available system resources:

Download the model from HuggingFace: `pranav-pvnn/codellama-7b-python-ai-assistant-full-gguf`

2. Prompt Format

CRITICAL: Always use this exact prompt format for optimal results:

```

Instruction:

[Your instruction here]

Response:

```

The model has been specifically trained to recognize this format. Do not deviate from it.

3. Usage Patterns

#### For llama.cpp:

```bash

./llama-cli -m codellama-7b-merged-Q4_K_M.gguf -p "### Instruction:\n[task description]\n### Response:\n"

```

#### For Python (llama-cpp-python):

```python

from llama_cpp import Llama

llm = Llama(model_path="codellama-7b-merged-Q4_K_M.gguf")

prompt = "### Instruction:\n[task description]\n### Response:\n"

output = llm(prompt, max_tokens=256)

print(output['choices'][0]['text'])

```

#### For Ollama:

1. Create a Modelfile:

```

FROM ./codellama-7b-merged-Q4_K_M.gguf

```

2. Create and run the model:

```bash

ollama create my-codellama -f Modelfile

ollama run my-codellama "[task description]"

```

4. Optimal Parameters

When generating code, use these recommended parameters:

5. Best Practices

1. **Be Specific**: Provide clear, detailed instructions about what the code should accomplish

2. **Include Context**: Mention any constraints (performance, libraries, Python version)

3. **Request Documentation**: Ask for docstrings and comments when needed

4. **Iterative Refinement**: For complex tasks, break them into smaller functions and build incrementally

5. **Validation**: Always review and test generated code before deployment

6. Task Categories

This model excels at:

7. Limitations

Example Usage

**Task**: Create a function to calculate factorial

**Prompt**:

```

Instruction:

Write a Python function to calculate factorial of a number with input validation and docstring.

Response:

```

**Expected Output**: A complete Python function with proper error handling, type hints, and documentation.

Technical Specifications

Notes