Instructions to use cs-552-2026-the-transformers/math_model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cs-552-2026-the-transformers/math_model with Transformers:

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

pipe = pipeline("text-generation", model="cs-552-2026-the-transformers/math_model")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cs-552-2026-the-transformers/math_model")
model = AutoModelForCausalLM.from_pretrained("cs-552-2026-the-transformers/math_model")
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 cs-552-2026-the-transformers/math_model with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cs-552-2026-the-transformers/math_model"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cs-552-2026-the-transformers/math_model",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cs-552-2026-the-transformers/math_model

SGLang

How to use cs-552-2026-the-transformers/math_model 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 "cs-552-2026-the-transformers/math_model" \
    --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": "cs-552-2026-the-transformers/math_model",
		"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 "cs-552-2026-the-transformers/math_model" \
        --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": "cs-552-2026-the-transformers/math_model",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cs-552-2026-the-transformers/math_model with Docker Model Runner:
```
docker model run hf.co/cs-552-2026-the-transformers/math_model
```

math_model

Commit History

Update Automated MNLP evaluation report (2026-06-01)

c00bca8
verified

zechen-nlp commited on 4 days ago

Automated MNLP evaluation report (2026-06-01)

ddbb7a5
verified

zechen-nlp commited on 4 days ago

exp6: SFT(NuminaMath) -> DPO v2, T=0.3

5184a78
verified

jdecim commited on 6 days ago

Automated MNLP evaluation report (2026-05-29) (#3)

fab7629

jdecim

zechen-nlp commited on 7 days ago

sft_mixlong_full: best Hard100 p@8, temperature=0.3

a421196
verified

jdecim commited on 7 days ago

Push DPO checkpoint with T=0.3 (optimal for pass@8 on CI gate)

ac3517e
verified

jdecim commited on 8 days ago

Fix README: add base_model, fix model references

6af8d3a
verified

jdecim commited on 9 days ago

Fix tokenizer_config.json: is_local=false

362bfeb
verified

jdecim commited on 9 days ago

Automated MNLP evaluation report (2026-05-27) (#2)

4bd5047

jdecim

zechen-nlp commited on 9 days ago

Upload exp4b: SFT→DPO v2 (L3-5, ≤5/8, beta=0.1, lr=5e-7)

32ac537
verified

jdecim commited on 10 days ago

Automated MNLP evaluation report (2026-05-23) (#1)

b8bbbde

NicolasBerlin

zechen-nlp commited on 12 days ago

Upload math SFT checkpoint

f13b0a9
verified

jdecim commited on 19 days ago

Upload math SFT checkpoint (smoke test)

ed23471
verified

jdecim commited on 22 days ago

Upload math SFT checkpoint (smoke test)

2b172d7
verified

jdecim commited on 25 days ago

initial commit

da27515
verified

jdecim commited on 26 days ago

Commit History

Update Automated MNLP evaluation report (2026-06-01) c00bca8 verified

Automated MNLP evaluation report (2026-06-01) ddbb7a5 verified

exp6: SFT(NuminaMath) -> DPO v2, T=0.3 5184a78 verified

Automated MNLP evaluation report (2026-05-29) (#3) fab7629

sft_mixlong_full: best Hard100 p@8, temperature=0.3 a421196 verified

Push DPO checkpoint with T=0.3 (optimal for pass@8 on CI gate) ac3517e verified

Fix README: add base_model, fix model references 6af8d3a verified

Fix tokenizer_config.json: is_local=false 362bfeb verified

Automated MNLP evaluation report (2026-05-27) (#2) 4bd5047

Upload exp4b: SFT→DPO v2 (L3-5, ≤5/8, beta=0.1, lr=5e-7) 32ac537 verified

Automated MNLP evaluation report (2026-05-23) (#1) b8bbbde

Upload math SFT checkpoint f13b0a9 verified

Upload math SFT checkpoint (smoke test) ed23471 verified

Upload math SFT checkpoint (smoke test) 2b172d7 verified

initial commit da27515 verified