Downloads · 30 days
540
2% of all-time downloads
vincent-pro-ai/french-llm-from-scratch-GGUF
french-llm-from-scratch-GGUF is a machine learning model from vincent-pro-ai. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Modèle francophone (~260M paramètres) entraîné from scratch. Compatible Mistral architecture, tokenizer Mistral 117k. Entraînement 100k steps sur conversations FR + FineWeb.
Downloads · 30 days
540
2% of all-time downloads
All-time downloads
30.6K
Public
Parameters
124M
11.7 GB on disk
Likes
1
Public
Click a slice to open those files.
.gguf4.7 GB · 63%
From the Hugging Face model README
Modèle francophone (~260M paramètres) entraîné from scratch. Compatible Mistral architecture, tokenizer Mistral 117k. Entraînement 100k steps sur conversations FR + FineWeb.
french-llm-v3-mistral-Q4_K_M.gguf (168 MB, quantisé Q4_K_M) ⭐french-llm-v3-mistral-f16.gguf (520 MB, précision float16)config.json, pytorch_model.bin (519 MB)tokenizer.json + config (vocabulaire Mistral 32k)vincent-pro-ai/french-llm-from-scratch-GGUFfrench-llm-v3-mistral-Q4_K_M.gguf (168 MB)pip install "transformers>=4.41" torch accelerate
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
repo = "vincent-pro-ai/french-llm-from-scratch-GGUF"
tokenizer = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(repo, torch_dtype=torch.float16, device_map="auto")
prompt = "Bonjour, comment ça va ?\n"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=100, temperature=0.8, top_p=0.9)
print(tokenizer.decode(out[0], skip_special_tokens=True))
MIT (voir dépôt source)