Instructions to use salmannyu/step_cot with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use salmannyu/step_cot with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("salmannyu/step_cot")
model = AutoModelForMultimodalLM.from_pretrained("salmannyu/step_cot")
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 salmannyu/step_cot with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/salmannyu/step_cot

SGLang

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

Docker Model Runner
How to use salmannyu/step_cot with Docker Model Runner:
```
docker model run hf.co/salmannyu/step_cot
```

step_cot / README.md

salmannyu

Upload folder using huggingface_hub

fa1352e verified 9 months ago

preview code

raw

history blame contribute delete

2.1 kB

	---
	library_name: transformers
	license: other
	base_model: Qwen/Qwen2.5-14B-Instruct
	tags:
	- llama-factory
	- full
	- generated_from_trainer
	model-index:
	- name: prm_cot
	results: []
	---

	<!-- This model card has been generated automatically according to the information the Trainer had access to. You
	should probably proofread and complete it, then remove this comment. -->

	# prm_cot

	This model is a fine-tuned version of [Qwen/Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) on the sky_math_step_level_gen_prm_cot dataset.
	It achieves the following results on the evaluation set:
	- Loss: 0.3308

	## Model description

	More information needed

	## Intended uses & limitations

	More information needed

	## Training and evaluation data

	More information needed

	## Training procedure

	### Training hyperparameters

	The following hyperparameters were used during training:
	- learning_rate: 5e-06
	- train_batch_size: 4
	- eval_batch_size: 2
	- seed: 42
	- distributed_type: multi-GPU
	- num_devices: 8
	- total_train_batch_size: 32
	- total_eval_batch_size: 16
	- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
	- lr_scheduler_type: cosine
	- lr_scheduler_warmup_ratio: 0.1
	- num_epochs: 3.0

	### Training results

	\| Training Loss \| Epoch \| Step \| Validation Loss \|
	\|:-------------:\|:------:\|:----:\|:---------------:\|
	\| 0.3221 \| 0.2539 \| 500 \| 0.3276 \|
	\| 0.3095 \| 0.5079 \| 1000 \| 0.3185 \|
	\| 0.308 \| 0.7618 \| 1500 \| 0.3138 \|
	\| 0.2496 \| 1.0157 \| 2000 \| 0.3163 \|
	\| 0.248 \| 1.2697 \| 2500 \| 0.3150 \|
	\| 0.2526 \| 1.5236 \| 3000 \| 0.3125 \|
	\| 0.254 \| 1.7776 \| 3500 \| 0.3094 \|
	\| 0.1867 \| 2.0315 \| 4000 \| 0.3303 \|
	\| 0.1863 \| 2.2854 \| 4500 \| 0.3317 \|
	\| 0.1825 \| 2.5394 \| 5000 \| 0.3310 \|
	\| 0.1787 \| 2.7933 \| 5500 \| 0.3310 \|


	### Framework versions

	- Transformers 4.51.1
	- Pytorch 2.6.0+cu124
	- Datasets 3.6.0
	- Tokenizers 0.21.1