Instructions to use MrOvkill/Phi-3-Instruct-Bloated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MrOvkill/Phi-3-Instruct-Bloated with Transformers:

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

pipe = pipeline("text-generation", model="MrOvkill/Phi-3-Instruct-Bloated", 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("MrOvkill/Phi-3-Instruct-Bloated", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("MrOvkill/Phi-3-Instruct-Bloated", trust_remote_code=True, device_map="auto")
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 MrOvkill/Phi-3-Instruct-Bloated with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/MrOvkill/Phi-3-Instruct-Bloated

SGLang

How to use MrOvkill/Phi-3-Instruct-Bloated 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 "MrOvkill/Phi-3-Instruct-Bloated" \
    --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": "MrOvkill/Phi-3-Instruct-Bloated",
		"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 "MrOvkill/Phi-3-Instruct-Bloated" \
        --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": "MrOvkill/Phi-3-Instruct-Bloated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MrOvkill/Phi-3-Instruct-Bloated with Docker Model Runner:
```
docker model run hf.co/MrOvkill/Phi-3-Instruct-Bloated
```

Phi-3-Instruct-Bloated / handler.py

MrOvkill

Update handler.py

dc0013e verified over 2 years ago

Raw

History Blame Contribute Delete

1.45 kB

	import json
	import os
	from typing import Dict, List, Any
	import torch
	from transformers import pipeline

	PROMPT_FORMAT= """
	<\|user\|>
	{inputs} <\|end\|>
	<\|assistant\|>
	"""

	class EndpointHandler():
	def __init__(self, data):
	cfg = {
	"repo": "MrOvkill/Phi-3-Instruct-Bloated",
	}
	self.pipe = pipeline("text-generation", "MrOvkill/Phi-3-Instruct-Bloated", torch_dtype=torch.float16, trust_remote_code=True)

	def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
	"""
	data args:
	inputs (:obj: `str` \| `PIL.Image` \| `np.array`)
	kwargs
	Return:
	A :obj:`list` \| `dict`: will be serialized and returned
	"""
	self.pipe = pipeline("text-generation", "MrOvkill/Phi-3-Instruct-Bloated", torch_dtype=torch.float16, trust_remote_code=True)
	max_new_tokens = 1024
	if "max_new_tokens" in data:
	max_new_tokens = data["max_new_tokens"]
	try:
	max_new_tokens = int(max_new_tokens)
	except Exception as e:
	return json.dumps({
	"status": "error",
	"reason": "max_length was passed as something that was absolutely not a plain old int"
	})

	res = PROMPT_FORMAT.format(inputs=data['inputs'])
	return self.pipe(
	res,
	do_sample=False,
	max_new_tokens=max_new_tokens
	)

	return res