Instructions to use mrs83/Kurtis-SmolLM2-1.7B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mrs83/Kurtis-SmolLM2-1.7B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="mrs83/Kurtis-SmolLM2-1.7B-Instruct")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mrs83/Kurtis-SmolLM2-1.7B-Instruct")
model = AutoModelForCausalLM.from_pretrained("mrs83/Kurtis-SmolLM2-1.7B-Instruct")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mrs83/Kurtis-SmolLM2-1.7B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mrs83/Kurtis-SmolLM2-1.7B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mrs83/Kurtis-SmolLM2-1.7B-Instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mrs83/Kurtis-SmolLM2-1.7B-Instruct

SGLang

How to use mrs83/Kurtis-SmolLM2-1.7B-Instruct 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 "mrs83/Kurtis-SmolLM2-1.7B-Instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mrs83/Kurtis-SmolLM2-1.7B-Instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "mrs83/Kurtis-SmolLM2-1.7B-Instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mrs83/Kurtis-SmolLM2-1.7B-Instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mrs83/Kurtis-SmolLM2-1.7B-Instruct with Docker Model Runner:
```
docker model run hf.co/mrs83/Kurtis-SmolLM2-1.7B-Instruct
```

Model Card for Kurtis-SmolLM2-1.7B-Instruct

This model has been fine-tuned using Kurtis, an experimental fine-tuning, inference and evaluation tool for Small Language Models.

Model Details

Model Description

Developed by: Massimo R. Scamarcia massimo.scamarcia@gmail.com
Funded by: Massimo R. Scamarcia massimo.scamarcia@gmail.com - (self-funded)
Shared by: Massimo R. Scamarcia massimo.scamarcia@gmail.com
Model type: Transformer decoder
Language(s) (NLP): English
License: MIT
Finetuned from model: HuggingFaceTB/SmolLM2-1.7B-Instruct

Model Sources

Repository: https://github.com/mrs83/kurtis

Uses

The model is intended for use in a conversational setting, particularly in mental health and therapeutic support scenarios.

Direct Use

Not suitable for production usage.

Out-of-Scope Use

This model should not be used for:

Making critical mental health decisions or diagnoses.
Replacing professional mental health services.
Applications where responses require regulatory compliance or are highly sensitive.
Generating responses without human supervision, especially in contexts that involve vulnerable individuals.

Bias, Risks, and Limitations

Misuse of this dataset could lead to providing inappropriate or harmful responses, so it should not be deployed without proper safeguards in place.

Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model.

How to Get Started with the Model

WIP

Downloads last month: 3

Safetensors

Model size

2B params

Tensor type

F32

BF16

Model tree for mrs83/Kurtis-SmolLM2-1.7B-Instruct

Base model

HuggingFaceTB/SmolLM2-1.7B

Quantized

HuggingFaceTB/SmolLM2-1.7B-Instruct

Quantized

(88)

this model

Dataset used to train mrs83/Kurtis-SmolLM2-1.7B-Instruct

Collection including mrs83/Kurtis-SmolLM2-1.7B-Instruct

Kurtis

Collection

Kurtis is a project focused on supervised fine-tuning of small language model (SLM) on mental-health datasets. • 8 items • Updated Jan 26, 2025