Instructions to use auchoi/phi3_4bit_merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use auchoi/phi3_4bit_merged with Transformers:

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

pipe = pipeline("text-generation", model="auchoi/phi3_4bit_merged")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("auchoi/phi3_4bit_merged")
model = AutoModelForCausalLM.from_pretrained("auchoi/phi3_4bit_merged", 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 auchoi/phi3_4bit_merged with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "auchoi/phi3_4bit_merged"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "auchoi/phi3_4bit_merged",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/auchoi/phi3_4bit_merged

SGLang

How to use auchoi/phi3_4bit_merged 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 "auchoi/phi3_4bit_merged" \
    --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": "auchoi/phi3_4bit_merged",
		"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 "auchoi/phi3_4bit_merged" \
        --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": "auchoi/phi3_4bit_merged",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use auchoi/phi3_4bit_merged with Docker Model Runner:
```
docker model run hf.co/auchoi/phi3_4bit_merged
```

Uploaded model

Developed by: auchoi
License: apache-2.0
Finetuned from model : unsloth/Phi-3-mini-4k-instruct-bnb-4bit

This was more of an experiment for me but this was an existing 4bit model that was quantized again to 4bit. it will not perform well

This mistral model was trained 2x faster with Unsloth and Huggingface's TRL library.

Downloads last month: 7

Safetensors

Model size

4B params

Tensor type

F32

BF16

Model tree for auchoi/phi3_4bit_merged

Base model

unsloth/Phi-3-mini-4k-instruct-bnb-4bit

Quantized

(65)

this model