Instructions to use SveBorg/gemmpen-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SveBorg/gemmpen-lora with Transformers:

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

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

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("SveBorg/gemmpen-lora", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use SveBorg/gemmpen-lora with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/SveBorg/gemmpen-lora

SGLang

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

Unsloth Studio

How to use SveBorg/gemmpen-lora 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 SveBorg/gemmpen-lora 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 SveBorg/gemmpen-lora to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for SveBorg/gemmpen-lora to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="SveBorg/gemmpen-lora",
    max_seq_length=2048,
)

Docker Model Runner
How to use SveBorg/gemmpen-lora with Docker Model Runner:
```
docker model run hf.co/SveBorg/gemmpen-lora
```

Update README.md

by DaisyGotThis - opened May 16

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

-3

Files changed (1) hide show

README.md +4 -3

README.md CHANGED Viewed

@@ -57,7 +57,7 @@ This LoRA adapter is the core of GemmPen. It was fine-tuned on real handwritten
 | | |
 |---|---|
 | **Live Demo** | [gemmpen.vercel.app](https://gemmpen.vercel.app) |
-| **Video** | [YouTube](https://youtu.be/u9QC1BEGuHQ) |
 | **Track** | Future of Education |
 | **Hackathon** | [The Gemma 4 Good Hackathon](https://www.kaggle.com/competitions/gemma-4-good-hackathon) |
@@ -211,7 +211,8 @@ This adapter is released under the Apache 2.0 license, following the base model'
 ## Links
 - [Live Demo](https://gemmpen.vercel.app) - Full web app with three example students
-- [Submission Video](https://youtu.be/u9QC1BEGuHQ) - 3-minute project walkthrough
 - [Hackathon Page](https://www.kaggle.com/competitions/gemma-4-good-hackathon)
 ## Citation
@@ -227,4 +228,4 @@ This adapter is released under the Apache 2.0 license, following the base model'
 ---
-*This model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth).*

 | | |
 |---|---|
 | **Live Demo** | [gemmpen.vercel.app](https://gemmpen.vercel.app) |
+| **Video** | [YouTube](https://youtu.be/IeDM1mJ3J2M) |
 | **Track** | Future of Education |
 | **Hackathon** | [The Gemma 4 Good Hackathon](https://www.kaggle.com/competitions/gemma-4-good-hackathon) |
 ## Links
 - [Live Demo](https://gemmpen.vercel.app) - Full web app with three example students
+- [Submission Video](https://youtu.be/IeDM1mJ3J2M) - 3-minute project walkthrough
+- [GitHub](https://github.com/SvenjaBorgwardt/GemmPen) - Source code for the web app
 - [Hackathon Page](https://www.kaggle.com/competitions/gemma-4-good-hackathon)
 ## Citation
 ---
+*This model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth).*