Instructions to use ba144220/cs224r-default-project-ipo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ba144220/cs224r-default-project-ipo with Transformers:

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

pipe = pipeline("text-generation", model="ba144220/cs224r-default-project-ipo")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ba144220/cs224r-default-project-ipo")
model = AutoModelForCausalLM.from_pretrained("ba144220/cs224r-default-project-ipo")
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 ba144220/cs224r-default-project-ipo with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ba144220/cs224r-default-project-ipo"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ba144220/cs224r-default-project-ipo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ba144220/cs224r-default-project-ipo

SGLang

How to use ba144220/cs224r-default-project-ipo 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 "ba144220/cs224r-default-project-ipo" \
    --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": "ba144220/cs224r-default-project-ipo",
		"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 "ba144220/cs224r-default-project-ipo" \
        --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": "ba144220/cs224r-default-project-ipo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ba144220/cs224r-default-project-ipo with Docker Model Runner:
```
docker model run hf.co/ba144220/cs224r-default-project-ipo
```

cs224r-default-project-ipo

IPO (Inverse Preference Optimization) fine-tuned model for the Countdown arithmetic reasoning task, built on top of an SFT baseline. Trained as part of Stanford CS224R (Spring 2026).

Model Description

This model is preference-tuned using IPO on pairwise chosen/rejected completions for Countdown problems. Given a target number and a set of allowed numbers, the model produces chain-of-thought reasoning inside <think> tags and a final answer inside <answer> tags.

Training Details

Hyperparameter	Value
Base model	ba144220/cs224r-default-project-sft (SFT-tuned Qwen2.5-0.5B)
Dataset	asingh15/countdown_tasks_3to4-dpo
Loss type	IPO
Beta	0.1
Epochs	1
Learning rate	5e-6
LR schedule	Cosine with 5% warmup
Batch size	64 (gradient accumulation = 16)
Weight decay	0.01
Precision	bfloat16
Gradient checkpointing	Enabled
Hardware	1x NVIDIA H100 (Modal)
Max prompt length	512
Max response length	1024

Evaluation

Evaluated on asingh15/countdown_tasks_3to4 test split (40 prompts) using vLLM with temperature 0.6, top-k 20, top-p 0.95, sampling K=16 responses per prompt.

Metric	SFT Baseline	IPO (this model)
Average Score	0.3660	0.4080
Pass@1	0.30	0.375
Pass@16	0.75 (30/40)	0.75 (30/40)
Correct (score=1.0)	244/800	287/800

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("ba144220/cs224r-default-project-ipo")
tokenizer = AutoTokenizer.from_pretrained("ba144220/cs224r-default-project-ipo")

messages = [{"role": "user", "content": "Using the numbers [3, 4, 6, 8], create an equation that equals 24."}]
input_text = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
inputs = tokenizer(input_text, return_tensors="pt").to(model.device)

outputs = model.generate(**inputs, max_new_tokens=1024, temperature=0.6, top_k=20, top_p=0.95, do_sample=True)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

Limitations

Trained and evaluated only on the Countdown arithmetic task; not intended for general-purpose use.
Performance degrades on harder problems with more numbers or larger targets.
The 0.5B parameter size limits reasoning capacity compared to larger models.

Authors

Yuchi Hsu (yuchihsu@stanford.edu) and Ryan He (ryanhe@stanford.edu), Stanford CS224R Spring 2026.

Downloads last month: 16

Safetensors

Model size

0.5B params

Tensor type

BF16

Model tree for ba144220/cs224r-default-project-ipo

Base model

Qwen/Qwen2.5-0.5B

Finetuned

ba144220/cs224r-default-project-sft

Finetuned

(2)

this model

Dataset used to train ba144220/cs224r-default-project-ipo

Evaluation results

Average Score on Countdown Tasks 3-to-4
test set self-reported

0.408
Pass@16 on Countdown Tasks 3-to-4
test set self-reported

0.750