Instructions to use NewstaR/Morningstar-13b-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NewstaR/Morningstar-13b-hf with Transformers:

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

pipe = pipeline("text-generation", model="NewstaR/Morningstar-13b-hf")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("NewstaR/Morningstar-13b-hf")
model = AutoModelForCausalLM.from_pretrained("NewstaR/Morningstar-13b-hf")

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use NewstaR/Morningstar-13b-hf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NewstaR/Morningstar-13b-hf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NewstaR/Morningstar-13b-hf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/NewstaR/Morningstar-13b-hf

SGLang

How to use NewstaR/Morningstar-13b-hf 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 "NewstaR/Morningstar-13b-hf" \
    --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": "NewstaR/Morningstar-13b-hf",
		"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 "NewstaR/Morningstar-13b-hf" \
        --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": "NewstaR/Morningstar-13b-hf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use NewstaR/Morningstar-13b-hf with Docker Model Runner:
```
docker model run hf.co/NewstaR/Morningstar-13b-hf
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

MorningStar

Model	Average ⬆️	ARC	HellaSwag	MMLU	TruthfulQA
NewstaR/Morningstar-13b-hf 📑	59.93	59.04	81.93	54.63	44.12

Model Details

Model name: MorningStar
Model type: LLaMa 2 (13 billion parameters)

Intended Use

Text generation
Content creation
Conversational agent

Capabilities

MorningStar is optimized for natural language processing tasks like text generation and dialogue. It can produce fluent, coherent text across a variety of topics.

Limitations

May generate incorrect or nonsensical text
Lacks true language understanding
Potential for generating biased or unsafe content

Training Data

Details on MorningStar's training data are unavailable. It was likely trained on a large corpus of text data scraped from the internet.

Ethical Considerations

Large language models like MorningStar carry risks around bias, toxicity, and misinformation.
Model outputs should be monitored and filtered before use in real applications.
Avoid harmful or unethical prompts.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	50.48
ARC (25-shot)	59.04
HellaSwag (10-shot)	81.93
MMLU (5-shot)	54.63
TruthfulQA (0-shot)	44.12
Winogrande (5-shot)	74.51
GSM8K (5-shot)	15.24
DROP (3-shot)	23.87

Downloads last month: 1,091

Safetensors

Model size

13B params

Tensor type

F16

Model tree for NewstaR/Morningstar-13b-hf

Quantizations

2 models

NewstaR
/

Morningstar-13b-hf