Instructions to use neopolita/doc-to-lora-ministral-3b-2512 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use neopolita/doc-to-lora-ministral-3b-2512 with Transformers:

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

pipe = pipeline("text-generation", model="neopolita/doc-to-lora-ministral-3b-2512")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("neopolita/doc-to-lora-ministral-3b-2512", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use neopolita/doc-to-lora-ministral-3b-2512 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "neopolita/doc-to-lora-ministral-3b-2512"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "neopolita/doc-to-lora-ministral-3b-2512",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/neopolita/doc-to-lora-ministral-3b-2512

SGLang

How to use neopolita/doc-to-lora-ministral-3b-2512 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 "neopolita/doc-to-lora-ministral-3b-2512" \
    --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": "neopolita/doc-to-lora-ministral-3b-2512",
		"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 "neopolita/doc-to-lora-ministral-3b-2512" \
        --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": "neopolita/doc-to-lora-ministral-3b-2512",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use neopolita/doc-to-lora-ministral-3b-2512 with Docker Model Runner:
```
docker model run hf.co/neopolita/doc-to-lora-ministral-3b-2512
```

neopolita commited on Mar 1

Commit

1f2a896

verified ·

1 Parent(s): 146cc93

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -125,7 +125,7 @@ This model is designed to be used with the [doc-to-lora](https://github.com/Neop
 - **Training code**: [Neopolita/doc-to-lora](https://github.com/Neopolita/doc-to-lora)
 - **Inference agent**: [Neopolita/thoth](https://github.com/Neopolita/thoth)
 - **Base model**: [mistralai/Ministral-3-3B-Instruct-2512](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512)
-- **Original paper**: [Doc-to-LoRA: Sub-Second Knowledge Injection into LLMs](https://pub.sakana.ai/doc-to-lora/) (Sakana AI, Feb 2026)
 - **Hackathon**: [Mistral AI Worldwide Hackathon 2026](https://worldwide-hackathon.mistral.ai/)
 ## Limitations

 - **Training code**: [Neopolita/doc-to-lora](https://github.com/Neopolita/doc-to-lora)
 - **Inference agent**: [Neopolita/thoth](https://github.com/Neopolita/thoth)
 - **Base model**: [mistralai/Ministral-3-3B-Instruct-2512](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512)
+- **Original paper**: [Instant LLM Updates with Doc-to-LoRA and Text-to-LoRA](https://pub.sakana.ai/doc-to-lora/) (Sakana AI, Feb 2026)
 - **Hackathon**: [Mistral AI Worldwide Hackathon 2026](https://worldwide-hackathon.mistral.ai/)
 ## Limitations