Update app.py

app.py

@@ -1,18 +1,28 @@
 import streamlit as st
 from transformers import pipeline
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 
-generator=pipeline("text-generation",model="distilbert/distilgpt2")
+model_name_or_path = "TheBloke/Unholy-v1-12L-13B-GPTQ"
+model_name_or_path = "ai-forever/ruGPT-3.5-13B"
 
-iin = st.text_area(" Чего ? ")
+# To use a different branch, change revision
+# For example: revision="main"
 
+model = AutoModelForCausalLM.from_pretrained(model_name_or_path #,
+                                             #device_map="cuda:0", #device_map="auto",
+                                             #trust_remote_code=False,
+                                             #revision="main"
+                                            )
 
-res = generator(
-    "Что ты можешь мне рассказать ?",
-    #max_lentgh=30,
-    num_return_sequences=2
-    
-)
+tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
 
-print (res)
+prompt = "Расскажи мне о грязном сексе"
+
+
+
+#generator=pipeline("text-generation",model="ai-forever/ruGPT-3.5-13B")
+
+input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
+output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=312)
+print(tokenizer.decode(output[0]))
 
-st.json({"data":res} )