Instructions to use AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2 with Transformers:

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

pipe = pipeline("text-generation", model="AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2")
model = AutoModelForCausalLM.from_pretrained("AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2", 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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2

SGLang

How to use AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2 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 "AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2" \
    --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": "AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2",
		"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 "AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2" \
        --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": "AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2 with Docker Model Runner:
```
docker model run hf.co/AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2
```

AmberYifan commited on Jun 25, 2025

Commit

881a6af

verified ·

1 Parent(s): 4f457ad

Model save

Browse files

Files changed (4) hide show

README.md +1 -1
all_results.json +4 -4
train_results.json +4 -4
trainer_state.json +0 -0

README.md CHANGED Viewed

@@ -27,7 +27,7 @@ print(output["generated_text"])
 ## 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/yifanwang/huggingface/runs/wzmnx19m)
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).

 ## 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/yifanwang/huggingface/runs/zjyyyz8s)
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).

all_results.json CHANGED Viewed

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 0.0,
-    "train_loss": 0.4912783460274646,
-    "train_runtime": 2179.3367,
     "train_samples": 10467,
-    "train_samples_per_second": 4.803,
-    "train_steps_per_second": 0.601
 }

 {
     "epoch": 1.0,
     "total_flos": 0.0,
+    "train_loss": 0.4910648308062572,
+    "train_runtime": 2162.833,
     "train_samples": 10467,
+    "train_samples_per_second": 4.839,
+    "train_steps_per_second": 0.605
 }

train_results.json CHANGED Viewed

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 0.0,
-    "train_loss": 0.4912783460274646,
-    "train_runtime": 2179.3367,
     "train_samples": 10467,
-    "train_samples_per_second": 4.803,
-    "train_steps_per_second": 0.601
 }

 {
     "epoch": 1.0,
     "total_flos": 0.0,
+    "train_loss": 0.4910648308062572,
+    "train_runtime": 2162.833,
     "train_samples": 10467,
+    "train_samples_per_second": 4.839,
+    "train_steps_per_second": 0.605
 }

trainer_state.json CHANGED Viewed

The diff for this file is too large to render. See raw diff