Upload folder using huggingface_hub

Dockerfile

@@ -25,8 +25,12 @@ ENV OLLAMA_LOAD_TIMEOUT="24h"
 RUN apt-get update && apt-get upgrade -y
 RUN apt-get install git g++ python3 python3-pip -y && apt-get clean
 
-COPY pull.sh pull.sh
-RUN /bin/bash -x pull.sh
+COPY pull06.sh pull06.sh
+RUN /bin/bash -x pull06.sh
+COPY pull8.sh pull8.sh
+RUN /bin/bash -x pull8.sh
+# COPY pull14.sh pull.sh
+# RUN /bin/bash -x pull14.sh
 
 COPY requirements.txt requirements.txt
 RUN pip install --no-cache-dir -r requirements.txt --break-system-packages

main.py

@@ -82,7 +82,34 @@ def test_time():
   return '\n'.join(result)
 
 
-@app.route('/time14')
+@app.route('/time8')
+def test_time8():
+  from  datetime import datetime
+  from ollama import Client
+  from ollama import chat
+
+  t_start=datetime.now()
+  ansver = []
+  messages = [
+  {
+    'role': 'user',
+    'content': 'Расскажи о себе подробно',
+  },
+  ]
+  response = chat('qwen3:8b', messages=messages)
+  ansver.append(f' start {t_start} ')
+  response_time=datetime.now()-t_start
+  ansver.append(f' duration  {response_time} ')
+  #
+  response_len = len(response['message']['content'])
+  ansver.append(f" lehgth {response_len}")
+  response_speed=response_len/int(response_time.seconds)
+  ansver.append(f" token/sek {response_len}")
+  ansver.append(response['message']['content'])
+  result = [f'<p>{answ}</p>' for answ in ansver ]
+  return '\n'.join(result)
+
+#@app.route('/time14')
 def test_time14():
   from  datetime import datetime
   from ollama import Client
@@ -96,7 +123,7 @@ def test_time14():
     'content': 'Расскажи о себе подробно',
   },
   ]
-  response = chat('qwen3:30b', messages=messages)
+  response = chat('qwen3:14b', messages=messages)
   ansver.append(f' start {t_start} ')
   response_time=datetime.now()-t_start
   ansver.append(f' duration  {response_time} ')
@@ -110,7 +137,7 @@ def test_time14():
   return '\n'.join(result)
 
 
-@app.route('/time30')
+#@app.route('/time30')
 def test_time30():
   from  datetime import datetime
   from ollama import Client

pull06.sh

@@ -0,0 +1,16 @@
+#/bin/bash
+while true 
+  do nohup ollama serve >/dev/null 2>&1  
+  echo  sleep 9
+  sleep 9
+  done >/dev/null 2>&1 & 
+echo  sleep 5
+sleep 5
+# /usr/bin/ollama pull qwen3-next:80b
+# /usr/bin/ollama pull qwen3-vl:235b
+/usr/bin/ollama pull qwen3:0.6b 
+/usr/bin/ollama pull qwen3:8b
+# /usr/bin/ollama pull qwen3:14b
+# /usr/bin/ollama pull qwen3:30b
+# /usr/bin/ollama pull qwen3.5:35b
+/usr/bin/ollama list

pull14.sh

@@ -0,0 +1,16 @@
+#/bin/bash
+while true 
+  do nohup ollama serve >/dev/null 2>&1  
+  echo  sleep 9
+  sleep 9
+  done >/dev/null 2>&1 & 
+echo  sleep 5
+sleep 5
+# /usr/bin/ollama pull qwen3-next:80b
+# /usr/bin/ollama pull qwen3-vl:235b
+# /usr/bin/ollama pull qwen3:0.6b
+# /usr/bin/ollama pull qwen3:8b
+/usr/bin/ollama pull qwen3:14b
+# /usr/bin/ollama pull qwen3:30b
+# /usr/bin/ollama pull qwen3.5:35b
+/usr/bin/ollama list

pull8.sh

@@ -0,0 +1,16 @@
+#/bin/bash
+while true 
+  do nohup ollama serve >/dev/null 2>&1  
+  echo  sleep 9
+  sleep 9
+  done >/dev/null 2>&1 & 
+echo  sleep 5
+sleep 5
+# /usr/bin/ollama pull qwen3-next:80b
+# /usr/bin/ollama pull qwen3-vl:235b
+/usr/bin/ollama pull qwen3:0.6b
+# /usr/bin/ollama pull qwen3:8b
+# /usr/bin/ollama pull qwen3:14b
+# /usr/bin/ollama pull qwen3:30b
+# /usr/bin/ollama pull qwen3.5:35b
+/usr/bin/ollama list

start.sh

@@ -1,11 +1,12 @@
 #/bin/bash
 while true 
   do nohup ollama serve >/dev/null 2>&1  
-  echo  sleep 9
-  sleep 9
+  echo  sleep 4
+  sleep 4
   done >/dev/null 2>&1 & 
-echo  sleep 30
-sleep 30
+echo  sleep 3
+sleep 3
+/bin/bash pull14.sh
 echo   list
 /usr/bin/ollama list
 python3 main.py