Instructions to use qgallouedec/xpo-qwen2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use qgallouedec/xpo-qwen2 with Transformers:

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

pipe = pipeline("text-generation", model="qgallouedec/xpo-qwen2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("qgallouedec/xpo-qwen2")
model = AutoModelForCausalLM.from_pretrained("qgallouedec/xpo-qwen2")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use qgallouedec/xpo-qwen2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "qgallouedec/xpo-qwen2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "qgallouedec/xpo-qwen2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/qgallouedec/xpo-qwen2

SGLang

How to use qgallouedec/xpo-qwen2 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 "qgallouedec/xpo-qwen2" \
    --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": "qgallouedec/xpo-qwen2",
		"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 "qgallouedec/xpo-qwen2" \
        --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": "qgallouedec/xpo-qwen2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use qgallouedec/xpo-qwen2 with Docker Model Runner:
```
docker model run hf.co/qgallouedec/xpo-qwen2
```

qgallouedec HF Staff commited on Sep 26, 2024

Commit

32b016f

verified ·

1 Parent(s): e44d545

End of training

Browse files

Files changed (1) hide show

README.md +6 -7

README.md CHANGED Viewed

@@ -28,7 +28,7 @@ print(output["generated_text"][1]["content"])
 ## Training procedure
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/huggingface/huggingface/runs/erixlr9e)
 This model was trained with XPO, a method introduced in [Exploratory Preference Optimization: Harnessing Implicit Q*-Approximation for Sample-Efficient RLHF](https://huggingface.co/papers/2405.21046).
@@ -46,12 +46,11 @@ Cite XPO as:
 ```bibtex
 @article{jung2024binary,
-            title        = {{Binary Classifier Optimization for Large Language Model Alignment}},
-            author       = {Seungjae Jung and Gunsoo Han and Daniel Wontae Nam and Kyoung{-}Woon On},
-            year         = 2024,
-            eprint       = {arXiv:2404.04656}
-        }
 ```
 Cite TRL as:

 ## Training procedure
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/huggingface/huggingface/runs/bg6y6mom)
 This model was trained with XPO, a method introduced in [Exploratory Preference Optimization: Harnessing Implicit Q*-Approximation for Sample-Efficient RLHF](https://huggingface.co/papers/2405.21046).
 ```bibtex
 @article{jung2024binary,
+    title        = {{Binary Classifier Optimization for Large Language Model Alignment}},
+    author       = {Seungjae Jung and Gunsoo Han and Daniel Wontae Nam and Kyoung{-}Woon On},
+    year         = 2024,
+    eprint       = {arXiv:2404.04656}
+}
 ```
 Cite TRL as: