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

Libraries

How to use cs-552-2026-databand/group_model with Transformers:

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

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

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cs-552-2026-databand/group_model")
model = AutoModelForCausalLM.from_pretrained("cs-552-2026-databand/group_model", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

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

Use Docker

docker model run hf.co/cs-552-2026-databand/group_model

SGLang

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

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

mmm128 commited on May 20

Commit

37823cd

verified ·

1 Parent(s): 5c9eeb3

Baseline: Qwen3-1.7B + thinking-ON chat template + group system prompt

Browse files

Files changed (1) hide show

chat_template.jinja +1 -1

chat_template.jinja CHANGED Viewed

@@ -2,7 +2,7 @@
    Prepended to the official Qwen/Qwen3-1.7B chat_template.jinja by
    scripts/push_baseline.py at deploy time. Applied to: math, group. #}
 {%- set enable_thinking = false -%}
-{%- set system_message = "Start your message with: final answer will be \\boxed{ANSWER}. Then, solve concisely. Do not ramble on, decompose the problem into subproblems. Do NOT TRY TO ANSWER BY TRYING THE CASE FOR EVERY INTEGER. If you find yourself doing that generate a final answer as \\boxed{NONE} and stop. Your final answer MUST BE IN A BOX as such \\boxed{ANSWER} only once. Do not say anything after giving your final answer" -%}
 {%- if tools %}
     {{- '<|im_start|>system\n' }}
     {%- if messages[0].role == 'system' %}

    Prepended to the official Qwen/Qwen3-1.7B chat_template.jinja by
    scripts/push_baseline.py at deploy time. Applied to: math, group. #}
 {%- set enable_thinking = false -%}
+{%- set system_message = "Start your message with: final answer will be \\boxed{ANSWER}. Then, solve concisely. Do not ramble on, decompose the problem into subproblems. Do NOT TRY TO ANSWER BY TRYING THE CASE FOR EVERY INTEGER IN MATH PROBLEMS. If you are given the choice between multiple options, generated your final answer as \\boxed{A} or \\boxed{B}. If you find yourself doing that generate a final answer as \\boxed{NONE} and stop. Your final answer MUST BE IN A BOX as such \\boxed{ANSWER} only once. Do not say anything after giving your final answer" -%}
 {%- if tools %}
     {{- '<|im_start|>system\n' }}
     {%- if messages[0].role == 'system' %}