Instructions to use xsanskarx/calculator-smollm2_v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xsanskarx/calculator-smollm2_v2 with Transformers:

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

pipe = pipeline("text-generation", model="xsanskarx/calculator-smollm2_v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("xsanskarx/calculator-smollm2_v2")
model = AutoModelForCausalLM.from_pretrained("xsanskarx/calculator-smollm2_v2")
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

vLLM

How to use xsanskarx/calculator-smollm2_v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "xsanskarx/calculator-smollm2_v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "xsanskarx/calculator-smollm2_v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/xsanskarx/calculator-smollm2_v2

SGLang

How to use xsanskarx/calculator-smollm2_v2 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 "xsanskarx/calculator-smollm2_v2" \
    --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": "xsanskarx/calculator-smollm2_v2",
		"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 "xsanskarx/calculator-smollm2_v2" \
        --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": "xsanskarx/calculator-smollm2_v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use xsanskarx/calculator-smollm2_v2 with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for xsanskarx/calculator-smollm2_v2 to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for xsanskarx/calculator-smollm2_v2 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for xsanskarx/calculator-smollm2_v2 to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="xsanskarx/calculator-smollm2_v2",
    max_seq_length=2048,
)

Docker Model Runner
How to use xsanskarx/calculator-smollm2_v2 with Docker Model Runner:
```
docker model run hf.co/xsanskarx/calculator-smollm2_v2
```

Uploaded model

Developed by: xsanskarx
License: apache-2.0
Finetuned from model : HuggingFaceTB/SmolLM2-360M-Instruct

license: apache-2.0 tags: - unsloth - trl - sft

Calculator-SMoL-LM2 v1 - LLM Calculator Experiment

Re finetuning it currenlty , this model is not that goo

previous v1calculator-smollm2_v1

🚨 Disclaimer 🚨

This model is a work in progress!

The current version (calculator-smollm2_v2) is an experimental model trained on a small subset of synthetic data (700,000 rows).
Dataset scaling is ongoing, and future versions will be trained on larger and more diverse datasets.
This model is not production-ready and is intended for educational and experimental purposes only.
The base model used is smollm2-360m-instruct from Hugging Face.

Use this model with caution, and do not rely on it for critical or real-world applications.

Model Description

Calculator-SMoL-LM2 v1 is a small language model fine-tuned on a synthetic dataset of arithmetic operations and comparisons. The model is designed to perform basic mathematical calculations and comparisons, such as addition, subtraction, multiplication, division, square roots, and equality checks. This is an experimental model trained on 700,000 rows of data generated using a custom script.

Intended Use

This model is intended for educational and experimental purposes. It can be used to:

Perform basic arithmetic operations (+, -, *, /).
Calculate square roots.
Compare numbers (>, <, =).

The model is not intended for real-world applications requiring high precision or reliability.

Limitations

Precision: The model is trained on synthetic data with limited precision (up to 5 decimal places). It may not handle very large numbers or highly precise calculations well.
Generalization: The model may struggle with inputs that deviate significantly from the training data distribution.
Error Handling: The model does not handle invalid inputs (e.g., division by zero) gracefully.
Scope: The model is limited to the operations it was trained on and may not generalize to more complex mathematical tasks.

Training Data

The model was trained on a synthetic dataset of 700,000 rows, generated using the following operations:

Addition (+)
Subtraction (-)
Multiplication (*)
Division (/)
Square Root (srqt)
Comparisons (=, >, <)

The dataset includes numbers with 3 to 10 digits in the integer part and up to 5 decimal places.

Downloads last month: 4

Model tree for xsanskarx/calculator-smollm2_v2

Base model

HuggingFaceTB/SmolLM2-360M

Quantized

HuggingFaceTB/SmolLM2-360M-Instruct

Finetuned

(143)

this model

xsanskarx
/

calculator-smollm2_v2