Instructions to use frameai/CodeLoxa-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use frameai/CodeLoxa-4B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="frameai/CodeLoxa-4B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("frameai/CodeLoxa-4B")
model = AutoModelForCausalLM.from_pretrained("frameai/CodeLoxa-4B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use frameai/CodeLoxa-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "frameai/CodeLoxa-4B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "frameai/CodeLoxa-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/frameai/CodeLoxa-4B

SGLang

How to use frameai/CodeLoxa-4B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "frameai/CodeLoxa-4B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "frameai/CodeLoxa-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "frameai/CodeLoxa-4B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "frameai/CodeLoxa-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use frameai/CodeLoxa-4B with Docker Model Runner:
```
docker model run hf.co/frameai/CodeLoxa-4B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

CodeLoxa-4B: Your Private Coding Partner 💻

Introduction

Welcome to CodeLoxa-4B, a cutting-edge, compact code generation model designed to revolutionize your private coding projects. Built with an extensive dataset of high-quality code, CodeLoxa-4B delivers exceptional accuracy and efficiency, making it an indispensable asset for developers seeking a reliable coding assistant.

Key Features

Optimized for Code Generation: CodeLoxa-4B specializes in understanding and generating code, ensuring that your development process is faster and more streamlined.
High Accuracy: With a remarkable 96% accuracy rate, this model guarantees the generation of precise and reliable code, reducing the need for extensive debugging and revisions.
Compact Size: Designed with efficiency in mind, CodeLoxa-4B's small size makes it perfect for integration into various environments without the need for extensive computational resources.
Private Project Focused: This model is tailored for use in private projects, offering a secure and efficient coding solution that respects the confidentiality of your work.

Performance

CodeLoxa-4B has been rigorously tested and optimized to achieve a 96% accuracy in code generation tasks. This high level of precision ensures that the model can handle a wide range of coding challenges, providing developers with a powerful tool to enhance their productivity.

Licensing

CodeLoxa-4B is available under a proprietary license. Please review the licensing terms to understand the usage rights and restrictions associated with this model. For detailed licensing information, please contact us at Community.

Getting Started

To start using CodeLoxa-4B in your projects, follow these simple steps:

from transformers import pipeline

generator = pipeline('text-generation', model='frameai/CodeLoxa-4B') 

text = generator("Write me a snake GUI game with points and loses metrics.", max_length=8192)
print(text[0]['generated_text'])

Example Applications

Automating Repetitive Coding Tasks: Quickly generate boilerplate code, freeing up developers to focus on more complex problems.
Code Refactoring and Optimization: Improve existing codebases by leveraging CodeLoxa-4B's ability to suggest more efficient code structures.
Educational Tool: Serve as an advanced learning aid for programmers, offering insights into coding best practices and syntax.

Support and Contact

For any questions, support requests, or feedback, please reach out to our support team at Community section. We are always here to help you get the most out of CodeLoxa-4B.

Contributing

We welcome contributions from the community! If you're interested in helping improve CodeLoxa-4B, please refer to contribution guidelines.

Acknowledgments

We extend our gratitude to the open-source community and all contributors who have played a role in the development of CodeLoxa-4B.

Disclaimer

CodeLoxa-4B is an AI-based code generation tool designed to assist with coding tasks. While it strives for high accuracy, users should review and test the generated code to ensure it meets their specific requirements and standards.

Downloads last month: 19

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for frameai/CodeLoxa-4B

Quantizations

2 models

Collection including frameai/CodeLoxa-4B

Loxa

Collection

a Loxa family models are best models to running on CPU and GPU with high quality(=>92% accuracy) • 5 items • Updated Feb 3, 2025 • 2