Instructions to use ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf with Transformers:

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

pipe = pipeline("text-generation", model="ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf", 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("ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf", 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 ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf

SGLang

How to use ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf 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 "ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf" \
    --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": "ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf",
		"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 "ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf" \
        --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": "ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf with Docker Model Runner:
```
docker model run hf.co/ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf
```

justheuristic commited on Jul 8, 2024

Commit

35b0e46

verified ·

1 Parent(s): b2d16a0

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -0

README.md CHANGED Viewed

@@ -25,6 +25,8 @@ Results:
 |  | [1x16g16 (1-bit, model link)](https://huggingface.co/ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf) | 7.42	| 10.40  | 2.7Gb |
 In general, we always recommend the 2-bit models for best accuracy-size trade-offs. If tempted to use the 1-bit model, try a smaller model ,
 e.g. Phi-3-**mini** quantized with AQLM+PV [(quantized model link)](https://huggingface.co/ISTA-DASLab/Phi-3-mini-4k-instruct-AQLM-PV-2Bit-1x16-hf) and compare the results, or check our [AQLM+PV collection](https://huggingface.co/collections/ISTA-DASLab/aqlmpv-66564dff5d84f00a893ba93f) for a more appropriate size.

 |  | [1x16g16 (1-bit, model link)](https://huggingface.co/ISTA-DASLab/Phi-3-medium-4k-instruct-AQLM-PV-1Bit-1x16-hf) | 7.42	| 10.40  | 2.7Gb |
+Phi-3-**medium** is not included in the original [PV-Tuining paper](https://arxiv.org/abs/2405.14852). As of yet, we did not have the bandwidth to evaluate it properly. We hope to eventually run the zero-shot evaluation suite, or you can help us by running it yourself and opening a pull-request to the readme!
 In general, we always recommend the 2-bit models for best accuracy-size trade-offs. If tempted to use the 1-bit model, try a smaller model ,
 e.g. Phi-3-**mini** quantized with AQLM+PV [(quantized model link)](https://huggingface.co/ISTA-DASLab/Phi-3-mini-4k-instruct-AQLM-PV-2Bit-1x16-hf) and compare the results, or check our [AQLM+PV collection](https://huggingface.co/collections/ISTA-DASLab/aqlmpv-66564dff5d84f00a893ba93f) for a more appropriate size.