app.py cambios gradio

app.py

@@ -1,11 +1,15 @@
 import os
 from threading import Thread
 from typing import Iterator
-
 import gradio as gr
 import spaces
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
+import huggingface_hub
+
+# Obtener token de Hugging Face
+token = os.environ.get("HUGGINGFACE_HUB_TOKEN", None)
+huggingface_hub.login(token=token)
 
 MAX_MAX_NEW_TOKENS = 2048
 DEFAULT_MAX_NEW_TOKENS = 1024
@@ -17,10 +21,10 @@ DESCRIPTION = """\
 
 This Space demonstrates model [Llama-2-13b-chat](https://huggingface.co/meta-llama/Llama-2-13b-chat) by Meta, a Llama 2 model with 13B parameters fine-tuned for chat instructions. Feel free to play with it, or duplicate to run generations without a queue! If you want to run your own service, you can also [deploy the model on Inference Endpoints](https://huggingface.co/inference-endpoints).
 
-🔎 For more details about the Llama 2 family of models and how to use them with `transformers`, take a look [at our blog post](https://huggingface.co/blog/llama2).
+ Para más detalles sobre la familia de modelos Llama 2 y cómo usarlos con `transformers`, echa un vistazo [a nuestro post de blog](https://huggingface.co/blog/llama2).
 
-🔨 Looking for an even more powerful model? Check out the large [**70B** model demo](https://huggingface.co/spaces/ysharma/Explore_llamav2_with_TGI).
-🐇 For a smaller model that you can run on many GPUs, check our [7B model demo](https://huggingface.co/spaces/huggingface-projects/llama-2-7b-chat).
+ Buscando un modelo aún más potente? ¡Echa un vistazo a la demo del modelo grande [**70B**](https://huggingface.co/spaces/ysharma/Explore_llamav2_with_TGI)!
+ Para un modelo más pequeño que puedas ejecutar en muchas GPU, echa un vistazo a nuestra [demo del modelo 7B](https://huggingface.co/spaces/huggingface-projects/llama-2-7b-chat).
 
 """
 
@@ -28,21 +32,25 @@ LICENSE = """
 <p/>
 
 ---
-As a derivate work of [Llama-2-13b-chat](https://huggingface.co/meta-llama/Llama-2-13b-chat) by Meta,
-this demo is governed by the original [license](https://huggingface.co/spaces/huggingface-projects/llama-2-13b-chat/blob/main/LICENSE.txt) and [acceptable use policy](https://huggingface.co/spaces/huggingface-projects/llama-2-13b-chat/blob/main/USE_POLICY.md).
+Como un trabajo derivado de [Llama-2-13b-chat](https://huggingface.co/meta-llama/Llama-2-13b-chat) de Meta,
+esta demo está gobernada por la [licencia original](https://huggingface.co/spaces/huggingface-projects/llama-2-13b-chat/blob/main/LICENSE.txt) y la [política de uso aceptable](https://huggingface.co/spaces/huggingface-projects/llama-2-13b-chat/blob/main/USE_POLICY.md).
 """
 
 if not torch.cuda.is_available():
-    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"
+    DESCRIPTION += "\n<p>Running on CPU. This demo does not work on CPU.</p>"
 
 
-if torch.cuda.is_available():
+model = None
+tokenizer = None
+
+@spaces.GPU
+def load_model():
+    global model, tokenizer
     model_id = "meta-llama/Llama-2-13b-chat-hf"
     model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_in_4bit=True)
     tokenizer = AutoTokenizer.from_pretrained(model_id)
     tokenizer.use_default_system_prompt = False
 
-
 @spaces.GPU
 def generate(
     message: str,
@@ -88,6 +96,9 @@ def generate(
         yield "".join(outputs)
 
 
+if gr.running:
+    load_model()
+
 chat_interface = gr.ChatInterface(
     fn=generate,
     additional_inputs=[
@@ -128,14 +139,6 @@ chat_interface = gr.ChatInterface(
             value=1.2,
         ),
     ],
-    stop_btn=None,
-    examples=[
-        ["Hello there! How are you doing?"],
-        ["Can you explain briefly to me what is the Python programming language?"],
-        ["Explain the plot of Cinderella in a sentence."],
-        ["How many hours does it take a man to eat a Helicopter?"],
-        ["Write a 100-word article on 'Benefits of Open-Source in AI research'"],
-    ],
     cache_examples=False,
 )
 

test.py

@@ -0,0 +1,32 @@
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import os
+import huggingface_hub
+
+# Obtener token de Hugging Face
+token = os.environ.get("HUGGINGFACE_HUB_TOKEN", None)
+huggingface_hub.login(token=token)
+
+# ID del modelo
+model_id = "CohereForAI/aya-23-35B"
+
+# Cargar tokenizador y modelo
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id)
+
+# Formatear mensaje con la plantilla de chat
+messages = [{"role": "user", "content": "Anneme onu ne kadar sevdiğimi anlatan bir mektup yaz"}]
+input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
+
+# Generar texto
+gen_tokens = model.generate(
+    input_ids,
+    max_new_tokens=100,
+    do_sample=True,
+    temperature=0.3,
+    force_download=True
+)
+
+# Decodificar tokens generados
+gen_text = tokenizer.decode(gen_tokens[0], skip_special_tokens=True)
+print(gen_text)
+