From scratch
Collection
Finisha SLM โจ โข 87 items โข Updated
How to use Finisha-F-scratch/ReeCi with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="Finisha-F-scratch/ReeCi")
messages = [
{"role": "user", "content": "Who are you?"},
]
pipe(messages) # Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("Finisha-F-scratch/ReeCi")
model = AutoModelForCausalLM.from_pretrained("Finisha-F-scratch/ReeCi")
messages = [
{"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
messages,
add_generation_prompt=True,
tokenize=True,
return_dict=True,
return_tensors="pt",
).to(model.device)
outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))How to use Finisha-F-scratch/ReeCi with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Finisha-F-scratch/ReeCi"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Finisha-F-scratch/ReeCi",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'docker model run hf.co/Finisha-F-scratch/ReeCi
How to use Finisha-F-scratch/ReeCi with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "Finisha-F-scratch/ReeCi" \
--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": "Finisha-F-scratch/ReeCi",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'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/ReeCi" \
--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": "Finisha-F-scratch/ReeCi",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'How to use Finisha-F-scratch/ReeCi with Docker Model Runner:
docker model run hf.co/Finisha-F-scratch/ReeCi
| Caractรฉristique | Dรฉtails |
|---|---|
| Nom du Projet | ReeCi |
| Crรฉateur | Clemylia ๐ |
| Base Model | Charlotte-Amity (51M Paramรจtres) ๐ |
| Statut de l'IA | ๐ฏ 100% Original (Fait Maison) |
| Rรดle | Gรฉnรฉration de Recettes de Cuisine Totalement Absurdes et Philosophique ๐คช |
ReeCi est spรฉcialisรฉ dans l'art de mรฉlanger la cuisine et l'existentialisme. Son style n'est pas seulement absurde, il est intentionnel et codifiรฉ :
Titre :, Ingrรฉdients :, Instructions : pour garantir la lisibilitรฉ du chaos.Loyautรฉ, Amour Melta, Erreur 403, Prรฉcipitation, Doute).tรฉtรฉquilibre, rรชveillance sรปler, Pรขteau des Flan-lamina), enrichissant son vocabulaire personnel.lamina, Melta, Charlotte-Amity, les Classes), agissant comme des รฉpices thรฉmatiques.ReeCi a รฉtรฉ fine-tunรฉ sur 220 exemples structurรฉs pour forcer la coexistence de la logique de format et de l'absurditรฉ du contenu.
| Domaine | Rรฉsultat | Note du Chef |
|---|---|---|
| Absurditรฉ | Maximale et Conceptuelle (mรฉlange code/รฉmotion) | โญโญโญโญโญ |
| Cohรฉrence Structurelle | Trรจs bonne (respect des entรชtes) | โญโญโญโญ |
| Gรฉnรฉration de Nรฉologismes | Capacitรฉ รฉlevรฉe et intentionnelle | ๐ |
รtant basรฉ sur un modรจle de 51M de paramรจtres, la performance en termes de longueur de texte (contexte) est optimisรฉe par les techniques de quantification.