Delete app.py
Browse files
app.py
DELETED
|
@@ -1,200 +0,0 @@
|
|
| 1 |
-
import streamlit as st
|
| 2 |
-
import subprocess
|
| 3 |
-
import threading
|
| 4 |
-
import time
|
| 5 |
-
import requests
|
| 6 |
-
import json
|
| 7 |
-
from typing import List, Dict
|
| 8 |
-
import sys
|
| 9 |
-
import os
|
| 10 |
-
|
| 11 |
-
API_KEYS = ["sk-adminkey02", "sk-testkey123", "sk-userkey456", "sk-demokey789"]
|
| 12 |
-
DEFAULT_API_KEY = "sk-adminkey02"
|
| 13 |
-
|
| 14 |
-
def start_fastapi():
|
| 15 |
-
try:
|
| 16 |
-
subprocess.Popen([
|
| 17 |
-
sys.executable, "-m", "uvicorn", "api:app",
|
| 18 |
-
"--host", "0.0.0.0",
|
| 19 |
-
"--port", "8000",
|
| 20 |
-
"--workers", "1"
|
| 21 |
-
])
|
| 22 |
-
print("🚀 FastAPI server starting on port 8000...")
|
| 23 |
-
except Exception as e:
|
| 24 |
-
st.error(f"Failed to start API server: {e}")
|
| 25 |
-
|
| 26 |
-
if 'api_started' not in st.session_state:
|
| 27 |
-
st.session_state.api_started = True
|
| 28 |
-
threading.Thread(target=start_fastapi, daemon=True).start()
|
| 29 |
-
time.sleep(8)
|
| 30 |
-
|
| 31 |
-
def call_api(messages: List[Dict], max_tokens: int = 512, temperature: float = 0.7, api_key: str = DEFAULT_API_KEY):
|
| 32 |
-
try:
|
| 33 |
-
response = requests.post(
|
| 34 |
-
"http://localhost:8000/v1/chat/completions",
|
| 35 |
-
headers={"Authorization": f"Bearer {api_key}"},
|
| 36 |
-
json={
|
| 37 |
-
"model": "zephyr-quiklang-3b-4k",
|
| 38 |
-
"messages": messages,
|
| 39 |
-
"max_tokens": max_tokens,
|
| 40 |
-
"temperature": temperature
|
| 41 |
-
},
|
| 42 |
-
timeout=120
|
| 43 |
-
)
|
| 44 |
-
|
| 45 |
-
if response.status_code == 200:
|
| 46 |
-
return response.json()
|
| 47 |
-
elif response.status_code == 401:
|
| 48 |
-
st.error("❌ Invalid API key")
|
| 49 |
-
return None
|
| 50 |
-
else:
|
| 51 |
-
st.error(f"API Error: {response.status_code} - {response.text}")
|
| 52 |
-
return None
|
| 53 |
-
|
| 54 |
-
except requests.exceptions.RequestException as e:
|
| 55 |
-
st.error(f"Failed to connect to API: {e}")
|
| 56 |
-
return None
|
| 57 |
-
|
| 58 |
-
def check_api_health():
|
| 59 |
-
try:
|
| 60 |
-
response = requests.get("http://localhost:8000/v1/health", timeout=10)
|
| 61 |
-
return response.status_code == 200
|
| 62 |
-
except:
|
| 63 |
-
return False
|
| 64 |
-
|
| 65 |
-
def get_api_info():
|
| 66 |
-
try:
|
| 67 |
-
response = requests.get("http://localhost:8000/v1", timeout=5)
|
| 68 |
-
if response.status_code == 200:
|
| 69 |
-
return response.json()
|
| 70 |
-
except:
|
| 71 |
-
pass
|
| 72 |
-
return None
|
| 73 |
-
|
| 74 |
-
def main():
|
| 75 |
-
st.set_page_config(
|
| 76 |
-
page_title="🧠 Zephyr Quiklang 3B-4K Chat",
|
| 77 |
-
page_icon="🧠",
|
| 78 |
-
layout="wide",
|
| 79 |
-
initial_sidebar_state="expanded"
|
| 80 |
-
)
|
| 81 |
-
|
| 82 |
-
st.title("🧠 Zephyr Quiklang 3B-4K")
|
| 83 |
-
st.markdown("*3B parameter model with OpenAI-compatible API (Q4_K_M quantized)*")
|
| 84 |
-
|
| 85 |
-
space_url = st.text_input(
|
| 86 |
-
"🌐 Your Space URL",
|
| 87 |
-
placeholder="https://your-username-your-space.hf.space",
|
| 88 |
-
help="Enter your Hugging Face Space URL for API examples"
|
| 89 |
-
)
|
| 90 |
-
|
| 91 |
-
api_healthy = check_api_health()
|
| 92 |
-
|
| 93 |
-
col1, col2, col3 = st.columns([2, 1, 1])
|
| 94 |
-
with col1:
|
| 95 |
-
if api_healthy:
|
| 96 |
-
st.success("✅ API Server is running")
|
| 97 |
-
else:
|
| 98 |
-
st.warning("⏳ API Server is starting...")
|
| 99 |
-
|
| 100 |
-
with col2:
|
| 101 |
-
if st.button("🔄 Refresh"):
|
| 102 |
-
st.rerun()
|
| 103 |
-
|
| 104 |
-
with col3:
|
| 105 |
-
if st.button("📊 API Info"):
|
| 106 |
-
info = get_api_info()
|
| 107 |
-
if info:
|
| 108 |
-
st.json(info)
|
| 109 |
-
|
| 110 |
-
if not api_healthy:
|
| 111 |
-
st.info("🕐 The model is loading in the background. First startup may take 2-3 minutes.")
|
| 112 |
-
st.stop()
|
| 113 |
-
|
| 114 |
-
with st.sidebar:
|
| 115 |
-
st.header("⚙️ Settings")
|
| 116 |
-
selected_api_key = st.selectbox("🔑 API Key", API_KEYS, index=0)
|
| 117 |
-
max_tokens = st.slider("Max Tokens", 50, 1024, 512)
|
| 118 |
-
temperature = st.slider("Temperature", 0.0, 1.0, 0.7, 0.1)
|
| 119 |
-
|
| 120 |
-
st.divider()
|
| 121 |
-
st.header("📡 API Access")
|
| 122 |
-
st.markdown(f"**Base URL:** `{space_url or 'https://your-space.hf.space'}/v1`")
|
| 123 |
-
st.code("/v1/models")
|
| 124 |
-
st.code("/v1/chat/completions")
|
| 125 |
-
st.code("/v1/health")
|
| 126 |
-
st.code("/v1/docs")
|
| 127 |
-
st.markdown("**Authentication Header:**")
|
| 128 |
-
st.code(f"Authorization: Bearer {selected_api_key}")
|
| 129 |
-
|
| 130 |
-
tab1, tab2 = st.tabs(["💬 Chat", "🔗 Code Examples"])
|
| 131 |
-
|
| 132 |
-
with tab1:
|
| 133 |
-
if "messages" not in st.session_state:
|
| 134 |
-
st.session_state.messages = []
|
| 135 |
-
|
| 136 |
-
for message in st.session_state.messages:
|
| 137 |
-
with st.chat_message(message["role"]):
|
| 138 |
-
st.markdown(message["content"])
|
| 139 |
-
if "stats" in message:
|
| 140 |
-
st.caption(message["stats"])
|
| 141 |
-
|
| 142 |
-
if prompt := st.chat_input("Ask me anything..."):
|
| 143 |
-
st.session_state.messages.append({"role": "user", "content": prompt})
|
| 144 |
-
with st.chat_message("user"):
|
| 145 |
-
st.markdown(prompt)
|
| 146 |
-
|
| 147 |
-
with st.chat_message("assistant"):
|
| 148 |
-
with st.spinner("🤔 Thinking..."):
|
| 149 |
-
api_messages = [
|
| 150 |
-
{"role": msg["role"], "content": msg["content"]}
|
| 151 |
-
for msg in st.session_state.messages if "stats" not in msg
|
| 152 |
-
]
|
| 153 |
-
result = call_api(api_messages, max_tokens, temperature, selected_api_key)
|
| 154 |
-
if result and 'choices' in result:
|
| 155 |
-
response = result['choices'][0]['message']['content']
|
| 156 |
-
usage = result.get('usage', {})
|
| 157 |
-
completion_tokens = usage.get('completion_tokens', 0)
|
| 158 |
-
generation_time = usage.get('total_tokens', 0) / 4.0
|
| 159 |
-
stats = f"📊 {completion_tokens} tokens • ~{generation_time:.1f}s • API: {selected_api_key[:12]}..."
|
| 160 |
-
st.markdown(response)
|
| 161 |
-
st.caption(stats)
|
| 162 |
-
st.session_state.messages.append({
|
| 163 |
-
"role": "assistant",
|
| 164 |
-
"content": response,
|
| 165 |
-
"stats": stats
|
| 166 |
-
})
|
| 167 |
-
|
| 168 |
-
col1, col2 = st.columns(2)
|
| 169 |
-
with col1:
|
| 170 |
-
if st.button("🗑️ Clear Chat"):
|
| 171 |
-
st.session_state.messages = []
|
| 172 |
-
st.rerun()
|
| 173 |
-
with col2:
|
| 174 |
-
if st.button("💾 Export Chat"):
|
| 175 |
-
chat_json = json.dumps(st.session_state.messages, indent=2)
|
| 176 |
-
st.download_button("📥 Download JSON", chat_json, "chat_export.json", "application/json")
|
| 177 |
-
|
| 178 |
-
with tab2:
|
| 179 |
-
st.subheader("🐍 Python Example")
|
| 180 |
-
base_url = f"{space_url or 'https://your-space.hf.space'}/v1"
|
| 181 |
-
python_code = f"""from openai import OpenAI
|
| 182 |
-
|
| 183 |
-
client = OpenAI(
|
| 184 |
-
base_url="{base_url}",
|
| 185 |
-
api_key="{selected_api_key}"
|
| 186 |
-
)
|
| 187 |
-
|
| 188 |
-
response = client.chat.completions.create(
|
| 189 |
-
model="zephyr-quiklang-3b-4k",
|
| 190 |
-
messages=[{{"role": "user", "content": "Hello!"}}],
|
| 191 |
-
max_tokens=512,
|
| 192 |
-
temperature=0.7
|
| 193 |
-
)
|
| 194 |
-
|
| 195 |
-
print(response.choices[0].message.content)
|
| 196 |
-
"""
|
| 197 |
-
st.code(python_code, language="python")
|
| 198 |
-
|
| 199 |
-
if __name__ == "__main__":
|
| 200 |
-
main()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|