Instructions to use TroyDoesAI/Phi-3-Context-Obedient-RAG with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TroyDoesAI/Phi-3-Context-Obedient-RAG with Transformers:

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

pipe = pipeline("text-generation", model="TroyDoesAI/Phi-3-Context-Obedient-RAG", 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("TroyDoesAI/Phi-3-Context-Obedient-RAG", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("TroyDoesAI/Phi-3-Context-Obedient-RAG", 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

vLLM

How to use TroyDoesAI/Phi-3-Context-Obedient-RAG with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/TroyDoesAI/Phi-3-Context-Obedient-RAG

SGLang

How to use TroyDoesAI/Phi-3-Context-Obedient-RAG 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 "TroyDoesAI/Phi-3-Context-Obedient-RAG" \
    --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": "TroyDoesAI/Phi-3-Context-Obedient-RAG",
		"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 "TroyDoesAI/Phi-3-Context-Obedient-RAG" \
        --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": "TroyDoesAI/Phi-3-Context-Obedient-RAG",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TroyDoesAI/Phi-3-Context-Obedient-RAG with Docker Model Runner:
```
docker model run hf.co/TroyDoesAI/Phi-3-Context-Obedient-RAG
```

New Phi-3.5 model is here

by virtualspectre - opened Aug 21, 2024

Discussion

virtualspectre

Aug 21, 2024

https://huggingface.co/microsoft/Phi-3.5-mini-instruct

TroyDoesAI

Owner Aug 22, 2024

@virtualspectre have you tested the Phi-3.5 model for RAG, I hear its pretty good as is.

virtualspectre

Aug 23, 2024

Nope, Just tested it, it is nowhere near the accuracy of your aligned model.

TroyDoesAI

Owner Aug 26, 2024

Nope, Just tested it, it is nowhere near the accuracy of your aligned model.

I did some research, and I hear it has some problems with refusals from my inner circle, do you know if there is a good base model I could start with that might not have refusals? I am looking to train another model but havent been able to pick the next parameter count and what people actually want from a model.. Thank you for any feedback I really appreciate you. I am gonna finish this 8B model and quantize it down then my work station will be free.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment