Instructions to use m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA with Transformers:

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

pipe = pipeline("text-generation", model="m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA

SGLang

How to use m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA 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 "m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA" \
    --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": "m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA",
		"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 "m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA" \
        --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": "m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA with Docker Model Runner:
```
docker model run hf.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-ITA
```

m-polignano commited on Aug 11, 2025

Commit

67fea6c

verified ·

1 Parent(s): f86fada

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -44,7 +44,7 @@ The **NEXT** family includes **four models**:
 <hr>
 **Colab Demo:** [A100 - 40GB - Colab Notebook](https://colab.research.google.com/drive/1mhZLAdpOr3TRq-ZTG4XD52J98orHj_na?usp=sharing)<br>
-The Model run on a single GPU, 19.56GB of VRAM used during inference by using a 4bit Quantization.
 <hr>

 <hr>
 **Colab Demo:** [A100 - 40GB - Colab Notebook](https://colab.research.google.com/drive/1mhZLAdpOr3TRq-ZTG4XD52J98orHj_na?usp=sharing)<br>
+The Model runs on a single GPU, 19.56GB of VRAM used during inference by using a 4bit Quantization.
 <hr>