Instructions to use Nabbers1999/Mini-Llama-3B-Instruct-0124 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Nabbers1999/Mini-Llama-3B-Instruct-0124 with Transformers:

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

pipe = pipeline("text-generation", model="Nabbers1999/Mini-Llama-3B-Instruct-0124")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Nabbers1999/Mini-Llama-3B-Instruct-0124")
model = AutoModelForCausalLM.from_pretrained("Nabbers1999/Mini-Llama-3B-Instruct-0124")
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 Nabbers1999/Mini-Llama-3B-Instruct-0124 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Nabbers1999/Mini-Llama-3B-Instruct-0124"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Nabbers1999/Mini-Llama-3B-Instruct-0124",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Nabbers1999/Mini-Llama-3B-Instruct-0124

SGLang

How to use Nabbers1999/Mini-Llama-3B-Instruct-0124 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 "Nabbers1999/Mini-Llama-3B-Instruct-0124" \
    --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": "Nabbers1999/Mini-Llama-3B-Instruct-0124",
		"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 "Nabbers1999/Mini-Llama-3B-Instruct-0124" \
        --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": "Nabbers1999/Mini-Llama-3B-Instruct-0124",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Nabbers1999/Mini-Llama-3B-Instruct-0124 with Docker Model Runner:
```
docker model run hf.co/Nabbers1999/Mini-Llama-3B-Instruct-0124
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Mini-Llama 3B Instruct - 0124

My base pretrain model has undergone full fine-tuning on an additional 350M tokens using portions of Tulu 3 and Nvidia Nemotron instruct sets. It is rough but functionsl, and still needs DPO training to align it with human preferences.

For the base pretrain, see: Nabbers1999/Mini-Llama-3B-Base-0124

** Special note and edit 01/27 - Ministral 3 3B specifically uses tied_word_embeddings set to true in the config.json, which was present in my base fine tune. However after instruct training this model was saved with it set to false, likely due to the trainer confusing it for another model due to the llamaficiation. I have now fixed the flag in this model. If you train on this model be aware that this is something to watch for.