Instructions to use tommytracx/CodeLlama-Edge-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tommytracx/CodeLlama-Edge-1.5B with Transformers:

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

pipe = pipeline("text-generation", model="tommytracx/CodeLlama-Edge-1.5B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tommytracx/CodeLlama-Edge-1.5B")
model = AutoModelForCausalLM.from_pretrained("tommytracx/CodeLlama-Edge-1.5B")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use tommytracx/CodeLlama-Edge-1.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tommytracx/CodeLlama-Edge-1.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tommytracx/CodeLlama-Edge-1.5B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/tommytracx/CodeLlama-Edge-1.5B

SGLang

How to use tommytracx/CodeLlama-Edge-1.5B 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 "tommytracx/CodeLlama-Edge-1.5B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tommytracx/CodeLlama-Edge-1.5B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "tommytracx/CodeLlama-Edge-1.5B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tommytracx/CodeLlama-Edge-1.5B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use tommytracx/CodeLlama-Edge-1.5B with Docker Model Runner:
```
docker model run hf.co/tommytracx/CodeLlama-Edge-1.5B
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

CodeLlama-Edge-1.5B

CodeLlama-Edge-1.5B is an edge-optimized variant of the CodeLlama series, designed to run efficiently on mobile and embedded devices using quantized or distilled formats.

Model Description

Model Type: Causal Language Model
Base Model: CodeLlama
Optimizations: Quantization-aware training, pruning, and edge-device compatibility
Parameters: 1.5 Billion
Intended Use: On-device coding assistance, embedded systems, low-power environments

Features

Token-efficient for code generation
Ideal for IDEs, mobile apps, IoT dev tools
Low memory and compute footprint

Example Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tommytracx/CodeLlama-Edge-1.5B")
model = AutoModelForCausalLM.from_pretrained("tommytracx/CodeLlama-Edge-1.5B")

input_text = "def quicksort(arr):"
inputs = tokenizer(input_text, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=64)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))

License

Apache 2.0

Author

Maintained by: tommytracx

Downloads last month: -

GGUF

Hardware compatibility

4-bit