Finisha-F-scratch/Nacid
Viewer • Updated • 2.15k • 57
How to use Finisha-F-scratch/SMCLEM with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="Finisha-F-scratch/SMCLEM") # Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("Finisha-F-scratch/SMCLEM")
model = AutoModelForCausalLM.from_pretrained("Finisha-F-scratch/SMCLEM")How to use Finisha-F-scratch/SMCLEM with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Finisha-F-scratch/SMCLEM"
# 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/SMCLEM",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/Finisha-F-scratch/SMCLEM
How to use Finisha-F-scratch/SMCLEM with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "Finisha-F-scratch/SMCLEM" \
--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/SMCLEM",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'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/SMCLEM" \
--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/SMCLEM",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use Finisha-F-scratch/SMCLEM with Docker Model Runner:
docker model run hf.co/Finisha-F-scratch/SMCLEM
SMCLEM est un moteur de génération divergente conçu from scratch. Il privilégie la collision de concepts et la plasticité narrative sur la rigueur encyclopédique. C’est un modèle de texture, pas de vérité.
Le modèle fonctionne sur un axe de Sémantique Collisionnelle :
Ne pas utiliser pour : Calculs, faits historiques, conseils médicaux ou recettes de cuisine conventionnelles. Utiliser pour : Briser le blocage de l'écrivain, générer des dialogues d'IA déviantes, et explorer de nouvelles syntaxes.