Instructions to use orionai/firefly-0.1-phi-3-mini-4k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use orionai/firefly-0.1-phi-3-mini-4k with Transformers:

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

pipe = pipeline("text-generation", model="orionai/firefly-0.1-phi-3-mini-4k", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("orionai/firefly-0.1-phi-3-mini-4k", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("orionai/firefly-0.1-phi-3-mini-4k", trust_remote_code=True)
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 Settings

vLLM

How to use orionai/firefly-0.1-phi-3-mini-4k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "orionai/firefly-0.1-phi-3-mini-4k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "orionai/firefly-0.1-phi-3-mini-4k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/orionai/firefly-0.1-phi-3-mini-4k

SGLang

How to use orionai/firefly-0.1-phi-3-mini-4k 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 "orionai/firefly-0.1-phi-3-mini-4k" \
    --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": "orionai/firefly-0.1-phi-3-mini-4k",
		"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 "orionai/firefly-0.1-phi-3-mini-4k" \
        --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": "orionai/firefly-0.1-phi-3-mini-4k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use orionai/firefly-0.1-phi-3-mini-4k with Docker Model Runner:
```
docker model run hf.co/orionai/firefly-0.1-phi-3-mini-4k
```

Model Card for Model ID

Firefly-0.1-phi-3-mini-4k is a finetuned version of microsoft/Phi-3-mini-4k-instruct, trained on a dataset of diverse prompts and responses generated by existing AI models which perform better than phi 3.

Model Description

While we have chosen to not make the training dataset available to the public right now, this decision may be reversed in the upcoming weeks, to make a better contribution to the open-source community.

Model Sources

Demo : As we currently do not have enough resources, we are unable to host a demo. However, if you feel that you are capable of doing so, you can put a link to your demo in the contributions tab and we may feature it here.

Downloads last month: 1

Safetensors

Model size

4B params

Tensor type

BF16