Instructions to use mrfakename/NeuralOrca-7B-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mrfakename/NeuralOrca-7B-v1 with Transformers:

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

pipe = pipeline("text-generation", model="mrfakename/NeuralOrca-7B-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mrfakename/NeuralOrca-7B-v1")
model = AutoModelForCausalLM.from_pretrained("mrfakename/NeuralOrca-7B-v1")
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 mrfakename/NeuralOrca-7B-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mrfakename/NeuralOrca-7B-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mrfakename/NeuralOrca-7B-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mrfakename/NeuralOrca-7B-v1

SGLang

How to use mrfakename/NeuralOrca-7B-v1 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 "mrfakename/NeuralOrca-7B-v1" \
    --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": "mrfakename/NeuralOrca-7B-v1",
		"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 "mrfakename/NeuralOrca-7B-v1" \
        --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": "mrfakename/NeuralOrca-7B-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mrfakename/NeuralOrca-7B-v1 with Docker Model Runner:
```
docker model run hf.co/mrfakename/NeuralOrca-7B-v1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

WARNING! This is a "Frankenmerge" - this model is untested!

NeuralOrca 7B V1

GGUF Models

By mrfakename

Please note that this is an experimental model. We cannot guarantee model quality.

This is the first (alpha) release of NeuralOrca. NeuralOrca is a merge of the following models:

mlabonne/NeuralHermes-2.5-Mistral-7B (This model is actually OpenHermes 2.5 finetuned on Intel's Neural Chat dataset and uses the ChatML prompt format, weight: 1.0)
Open-Orca/Mistral-7B-OpenOrca (This model uses the ChatML prompt format, weight: 0.7)

Prompt Format

We use the ChatML prompt format.

Example:

<|im_start|>system
You are NeuralOrca, a helpful AI assistant.
<|im_end|>
<|im_start|>user
How are you?<|im_end|>
<|im_start|>assistant

Evaluations

Coming soon

Context Length

The context length for this model is 8192 tokens (8K).

License

You are responsible for your use of NeuralOrca.

This software is licensed under the Apache 2.0 license. If you want to use it for commercial use, it's probably fine but please contact me first.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	67.64
AI2 Reasoning Challenge (25-Shot)	65.27
HellaSwag (10-Shot)	85.07
MMLU (5-Shot)	63.68
TruthfulQA (0-shot)	54.58
Winogrande (5-shot)	78.77
GSM8k (5-shot)	58.45