added OpenAI support and markdown in the chat window

local_rag.py

@@ -1,5 +1,5 @@
 """
-Local RAG setup with LangChain, Ollama, and FAISS
+Local RAG setup with LangChain, Ollama/OpenAI, and FAISS
 Minimal dependencies, simple code
 """
 import os
@@ -9,34 +9,51 @@ from langchain_community.document_loaders import PyPDFLoader, TextLoader
 from langchain_community.vectorstores import FAISS
 from langchain_huggingface import HuggingFaceEmbeddings
 from langchain_ollama import ChatOllama
+from langchain_openai import ChatOpenAI
 from langchain_text_splitters import RecursiveCharacterTextSplitter
 
 
 class LocalRAG:
-    def __init__(self, vectorstore_path="./vectorstore", ollama_model="mistral:7b"):
+    def __init__(
-        """Initialize local RAG system"""
+        self,
+        vectorstore_path="./vectorstore",
+        llm_provider="ollama",
+        ollama_model="gpt-oss:20b",
+        openai_model="gpt-5.2",
+        ollama_base_url="http://localhost:11434",
+    ):
+        """Initialize local RAG system. llm_provider: 'ollama' or 'openai'."""
         self.vectorstore_path = vectorstore_path
-        self.ollama_model = ollama_model
+        self.llm_provider = llm_provider
-        
+
         # Embeddings
         print("Loading embeddings model...")
         self.embeddings = HuggingFaceEmbeddings(
             model_name="sentence-transformers/all-MiniLM-L6-v2"
         )
-        
+
         # Text splitter
         self.text_splitter = RecursiveCharacterTextSplitter(
             chunk_size=2000,
             chunk_overlap=400
         )
-        
+
-        # Ollama LLM
+        # LLM (Ollama or OpenAI)
-        print(f"Connecting to Ollama (model: {ollama_model})...")
+        if llm_provider == "openai":
-        self.llm = ChatOllama(
+            api_key = os.environ.get("OPENAI_API_KEY")
-            model=ollama_model,
+            if not api_key:
-            base_url="http://localhost:11434"
+                raise ValueError(
-        )
+                    "OPENAI_API_KEY environment variable is required when llm_provider='openai'"
-        
+                )
+            print(f"Using OpenAI (model: {openai_model})...")
+            self.llm = ChatOpenAI(model=openai_model, api_key=api_key)
+        else:
+            print(f"Using Ollama (model: {ollama_model})...")
+            self.llm = ChatOllama(
+                model=ollama_model,
+                base_url=ollama_base_url
+            )
+
         # Vector store (load if exists, otherwise None)
         self.vectorstore = None
         self._load_vectorstore()

requirements.txt

@@ -1,6 +1,7 @@
 langchain
 langchain-community
 langchain-ollama
+langchain-openai
 langchain-text-splitters
 langchain-huggingface
 faiss-cpu

server.py

@@ -10,10 +10,18 @@ from pydantic import BaseModel
 
 from local_rag import LocalRAG
 
-# Initialize RAG once at startup
+# LLM provider: "ollama" or "openai"
-VECTORSTORE_PATH = "./vectorstore"
+LLM_PROVIDER = "openai"
 OLLAMA_MODEL = "gpt-oss:20b"
-rag = LocalRAG(vectorstore_path=VECTORSTORE_PATH, ollama_model=OLLAMA_MODEL)
+OPENAI_MODEL = "gpt-5.2"
+
+VECTORSTORE_PATH = "./vectorstore"
+rag = LocalRAG(
+    vectorstore_path=VECTORSTORE_PATH,
+    llm_provider=LLM_PROVIDER,
+    ollama_model=OLLAMA_MODEL,
+    openai_model=OPENAI_MODEL,
+)
 
 app = FastAPI(title="Local RAG Chat", version="1.0.0")
 
@@ -64,15 +72,16 @@ def chat(request: ChatRequest):
         answer = result["answer"]
         retrieved = result.get("retrieved", [])
 
-        # Server-side console trace: log retrieved chunks before LLM answer
+        # Server-side console trace: shorter chunk logs + raw LLM response
         if retrieved:
-            print(f"\n[RAG] Retrieved {len(retrieved)} chunk(s) for query: {request.message[:80]!r}")
+            print(f"\n[RAG] Retrieved {len(retrieved)} chunk(s)")
             for i, chunk in enumerate(retrieved):
                 content = chunk.get("content", "")
-                preview = (content[:1000] + "...") if len(content) > 1000 else content
+                preview = (content[:80] + "...") if len(content) > 80 else content
-                print(f"  [{i + 1}] source={chunk.get('source', '')} page={chunk.get('page')} | {preview!r}")
+                print(f"  [{i + 1}] {chunk.get('source', '')} p.{chunk.get('page', '?')} s={chunk.get('score')} | {preview!r}")
         else:
-            print(f"\n[RAG] Retrieved 0 chunks for query: {request.message[:80]!r}")
+            print(f"\n[RAG] Retrieved 0 chunks")
+        print(f"[RAG] LLM response:\n{answer}")
 
         return ChatResponse(answer=answer, retrieved=retrieved)
     except Exception as e:

templates/chat.html

@@ -45,9 +45,45 @@
       border-radius: 8px;
       font-size: 0.9rem;
       line-height: 1.5;
-      white-space: pre-wrap;
       word-break: break-word;
     }
+    .msg.user, .msg.error {
+      white-space: pre-wrap;
+    }
+    .msg.assistant .markdown-body {
+      white-space: normal;
+    }
+    .msg.assistant .markdown-body h1, .msg.assistant .markdown-body h2, .msg.assistant .markdown-body h3 {
+      margin: 0.75em 0 0.35em;
+      font-size: 1em;
+      font-weight: 600;
+    }
+    .msg.assistant .markdown-body h1:first-child, .msg.assistant .markdown-body h2:first-child, .msg.assistant .markdown-body h3:first-child { margin-top: 0; }
+    .msg.assistant .markdown-body p { margin: 0.5em 0; }
+    .msg.assistant .markdown-body p:first-child { margin-top: 0; }
+    .msg.assistant .markdown-body p:last-child { margin-bottom: 0; }
+    .msg.assistant .markdown-body pre {
+      margin: 0.5em 0;
+      padding: 0.6rem;
+      background: #18181b;
+      border-radius: 6px;
+      overflow-x: auto;
+      font-size: 0.85em;
+    }
+    .msg.assistant .markdown-body code {
+      background: #18181b;
+      padding: 0.15em 0.35em;
+      border-radius: 4px;
+      font-size: 0.9em;
+    }
+    .msg.assistant .markdown-body pre code {
+      padding: 0;
+      background: none;
+    }
+    .msg.assistant .markdown-body ul, .msg.assistant .markdown-body ol { margin: 0.5em 0; padding-left: 1.4em; }
+    .msg.assistant .markdown-body li { margin: 0.25em 0; }
+    .msg.assistant .markdown-body a { color: #60a5fa; text-decoration: none; }
+    .msg.assistant .markdown-body a:hover { text-decoration: underline; }
     .msg.user {
       align-self: flex-end;
       background: #3f3f46;
@@ -120,11 +156,13 @@
       color: #71717a;
     }
   </style>
+  <script src="https://cdn.jsdelivr.net/npm/marked/marked.min.js"></script>
+  <script src="https://cdn.jsdelivr.net/npm/dompurify/dist/purify.min.js"></script>
 </head>
 <body>
   <header>
     <h1>Local RAG Chat</h1>
-    <p>Ask questions about your documents. Answers are generated from the vector store + Ollama.</p>
+    <p>Ask questions about your documents. Answers are generated from the vector store + Ollama / OpenAI.</p>
   </header>
 
   <div id="messages"></div>
@@ -143,10 +181,18 @@
     const chatHistory = [];
 
     function appendMessage(role, text, isError = false) {
+      text = text ?? '';
       const div = document.createElement('div');
       div.className = 'msg ' + (isError ? 'error' : role);
       const label = role === 'user' ? 'You' : 'RAG';
-      div.innerHTML = '<span class="label">' + label + '</span>' + escapeHtml(text);
+      let body;
+      if (role === 'assistant' && !isError) {
+        const rawHtml = marked.parse(text, { gfm: true, breaks: true });
+        body = '<div class="markdown-body">' + DOMPurify.sanitize(rawHtml) + '</div>';
+      } else {
+        body = escapeHtml(text);
+      }
+      div.innerHTML = '<span class="label">' + label + '</span>' + body;
       messagesEl.appendChild(div);
       messagesEl.scrollTop = messagesEl.scrollHeight;
       chatHistory.push({ role: role, content: text });