Instructions to use arnavgrg/phi-2-nf4-fp16-upscaled with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use arnavgrg/phi-2-nf4-fp16-upscaled with Transformers:

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

pipe = pipeline("text-generation", model="arnavgrg/phi-2-nf4-fp16-upscaled", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("arnavgrg/phi-2-nf4-fp16-upscaled", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("arnavgrg/phi-2-nf4-fp16-upscaled", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use arnavgrg/phi-2-nf4-fp16-upscaled with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "arnavgrg/phi-2-nf4-fp16-upscaled"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "arnavgrg/phi-2-nf4-fp16-upscaled",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/arnavgrg/phi-2-nf4-fp16-upscaled

SGLang

How to use arnavgrg/phi-2-nf4-fp16-upscaled 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 "arnavgrg/phi-2-nf4-fp16-upscaled" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "arnavgrg/phi-2-nf4-fp16-upscaled",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "arnavgrg/phi-2-nf4-fp16-upscaled" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "arnavgrg/phi-2-nf4-fp16-upscaled",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use arnavgrg/phi-2-nf4-fp16-upscaled with Docker Model Runner:
```
docker model run hf.co/arnavgrg/phi-2-nf4-fp16-upscaled
```

This is an upscaled fp16 variant of the original Phi-2 base model by Microsoft after it has been loaded with nf4 4-bit quantization via bitsandbytes. The main idea here is to upscale the linear4bit layers to fp16 so that the quantization/dequantization cost doesn't have to paid for each forward pass at inference time.

Note: The quantization operation to nf4 is not lossless, so the model weights for the linear layers are lossy, which means that this model will not work as well as the official base model.

To use this model, you can just load it via transformers in fp16:

import torch
from transformers import AutoModelForCausalLM

model = AutoModelForCausalLM.from_pretrained(
  "arnavgrg/phi-2-nf4-fp16-upscaled",
  device_map="auto",
  torch_dtype=torch.float16,
)

Downloads last month: 9

Safetensors

Model size

3B params

Tensor type

F16