File size: 6,821 Bytes
1ffeec5 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 | """
BeastBullet + RAG: Local Perplexity Clone
Build your own Perplexity for $0/month
"""
import requests
from typing import List, Dict
import json
from pathlib import Path
class WebRAG:
"""Web search and retrieval for BeastBullet."""
def __init__(self, search_engine="duckduckgo"):
self.search_engine = search_engine
def search_web(self, query: str, num_results: int = 5) -> List[Dict]:
"""Search the web and return results."""
# Using DuckDuckGo (free, no API key needed)
try:
from duckduckgo_search import DDGS
results = []
with DDGS() as ddgs:
for r in ddgs.text(query, max_results=num_results):
results.append({
"title": r.get("title", ""),
"url": r.get("href", ""),
"snippet": r.get("body", "")
})
return results
except ImportError:
print("⚠️ Install duckduckgo-search: pip install duckduckgo-search")
return []
def fetch_content(self, url: str) -> str:
"""Fetch full content from URL."""
try:
from bs4 import BeautifulSoup
response = requests.get(url, timeout=5)
soup = BeautifulSoup(response.content, 'html.parser')
# Extract main content
for script in soup(["script", "style"]):
script.decompose()
text = soup.get_text()
lines = (line.strip() for line in text.splitlines())
chunks = (phrase.strip() for line in lines for phrase in line.split(" "))
text = ' '.join(chunk for chunk in chunks if chunk)
return text[:5000] # Limit to 5000 chars
except:
return ""
def search_and_retrieve(self, query: str) -> Dict:
"""Search web and retrieve relevant content."""
print(f"🔍 Searching: {query}")
# Search
results = self.search_web(query)
if not results:
return {"sources": [], "context": ""}
# Fetch content from top results
sources = []
context_parts = []
for i, result in enumerate(results[:3], 1): # Top 3 results
print(f" 📄 Fetching: {result['title'][:50]}...")
content = self.fetch_content(result['url'])
if content:
sources.append({
"title": result['title'],
"url": result['url'],
"snippet": result['snippet']
})
context_parts.append(f"[Source {i}] {result['title']}\n{content[:1000]}")
return {
"sources": sources,
"context": "\n\n".join(context_parts)
}
class LocalPerplexity:
"""Local Perplexity clone using BeastBullet + RAG."""
def __init__(self):
from code.integration_example import TinyLlamaIntegration, ExpertLoader
print("🧠 Loading BeastBullet...")
loader = ExpertLoader("experts/")
loader.load_encoder()
loader.load_all_experts()
self.bb = TinyLlamaIntegration(
experts=loader.experts,
encoder=loader.encoder
)
self.rag = WebRAG()
print("✅ Local Perplexity ready!\n")
def query(self, question: str) -> Dict:
"""Query with web search + expert reasoning."""
print("=" * 70)
print(f"Question: {question}")
print("=" * 70)
# Step 1: Search and retrieve
retrieval = self.rag.search_and_retrieve(question)
if not retrieval["sources"]:
print("\n⚠️ No web results found, using BeastBullet knowledge only")
result = self.bb.query(question)
return result
# Step 2: Build context-aware query
context_query = f"""Based on the following sources, answer: {question}
Sources:
{retrieval['context']}
Provide a clear answer with citations."""
# Step 3: Query BeastBullet with context
result = self.bb.query(context_query)
# Step 4: Add sources to result
result["sources"] = retrieval["sources"]
return result
def display_result(self, result: Dict):
"""Display result in Perplexity-style format."""
print("\n" + "=" * 70)
print("🧠 Answer:")
print("=" * 70)
print(result.get("answer", "No answer generated"))
if "sources" in result and result["sources"]:
print("\n" + "=" * 70)
print("📚 Sources:")
print("=" * 70)
for i, source in enumerate(result["sources"], 1):
print(f"\n[{i}] {source['title']}")
print(f" {source['url']}")
print(f" {source['snippet'][:100]}...")
print("\n" + "=" * 70)
print("📊 Metadata:")
print("=" * 70)
print(f"Experts Used: {', '.join(result.get('experts_used', []))}")
print(f"Confidence: {result.get('confidence', 0):.1%}")
print("=" * 70)
# ============================================================================
# DEMO
# ============================================================================
def demo():
"""Run demo queries."""
perplexity = LocalPerplexity()
# Example queries
queries = [
"What are the latest AI developments in December 2025?",
"How does quantum entanglement work?",
"What is the current price of Bitcoin?",
]
for query in queries:
result = perplexity.query(query)
perplexity.display_result(result)
print("\n\n")
def interactive():
"""Interactive mode."""
perplexity = LocalPerplexity()
print("🧠 Local Perplexity - Interactive Mode")
print("=" * 70)
print("Ask any question. Type 'exit' to quit.")
print("=" * 70)
print()
while True:
try:
question = input("You: ").strip()
if question.lower() in ['exit', 'quit', 'q']:
print("\n👋 Goodbye!")
break
if not question:
continue
result = perplexity.query(question)
perplexity.display_result(result)
print("\n")
except KeyboardInterrupt:
print("\n\n👋 Goodbye!")
break
if __name__ == "__main__":
import sys
if len(sys.argv) > 1 and sys.argv[1] == "demo":
demo()
else:
interactive()
|