Instructions to use Finisha-F-scratch/Lamina-large-2b-pretrain with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Finisha-F-scratch/Lamina-large-2b-pretrain with Transformers:

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

pipe = pipeline("text-generation", model="Finisha-F-scratch/Lamina-large-2b-pretrain")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Finisha-F-scratch/Lamina-large-2b-pretrain")
model = AutoModelForCausalLM.from_pretrained("Finisha-F-scratch/Lamina-large-2b-pretrain", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Finisha-F-scratch/Lamina-large-2b-pretrain with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Finisha-F-scratch/Lamina-large-2b-pretrain"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Finisha-F-scratch/Lamina-large-2b-pretrain",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Finisha-F-scratch/Lamina-large-2b-pretrain

SGLang

How to use Finisha-F-scratch/Lamina-large-2b-pretrain 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 "Finisha-F-scratch/Lamina-large-2b-pretrain" \
    --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": "Finisha-F-scratch/Lamina-large-2b-pretrain",
		"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 "Finisha-F-scratch/Lamina-large-2b-pretrain" \
        --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": "Finisha-F-scratch/Lamina-large-2b-pretrain",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Finisha-F-scratch/Lamina-large-2b-pretrain with Docker Model Runner:
```
docker model run hf.co/Finisha-F-scratch/Lamina-large-2b-pretrain
```

Clemylia commited on Nov 24, 2025

Commit

879e1ce

verified ·

1 Parent(s): 0e03638

Update README.md

Browse files

Files changed (1) hide show

README.md +4 -1

README.md CHANGED Viewed

@@ -70,4 +70,7 @@ Ce modèle est la solution au problème rencontré avec **Lam-3** ($714$M) :
 ## 🤝 Remerciements
-Un grand merci aux soutiens de Clemylia, notamment à **Nora** et à sa famille, dont l'accès aux ressources de calcul TPU a rendu l'entraînement de ce modèle de $2$ milliards de paramètres **gratuit** et possible pour la communauté *open source*.

 ## 🤝 Remerciements
+Un grand merci aux soutiens de Clemylia, notamment à **Nora** et à sa famille, dont l'accès aux ressources de calcul TPU a rendu l'entraînement de ce modèle de $2$ milliards de paramètres **gratuit** et possible pour la communauté *open source*.
+🛑 **Ce modèle n'est pas utilisé en état,
+il est sous-entraînés, et son existence demontre juste la capacité d'entraîner un modèle aussi gros sur des ressources limitées**