How to Build a RAG Chatbot with DeepSeek API (Step-by-Step Guide 2026)

RAG (Retrieval-Augmented Generation) = Search + LLM.

Instead of relying on the LLM's training data (which may be outdated), RAG:
1. Indexes your documents (PDFs, web pages, manuals)
2. Retrieves the most relevant chunks when a user asks a question
3. Generates an answer using DeepSeek API with the retrieved context

import openai
client = openai.OpenAI(
    api_key="YOUR_AICREDITS_API_KEY",
    base_url="https://api.aicreditsapi.com/v1"
)

Get your API key from aicreditsapi.com/buy — no Chinese phone number required.

from sentence_transformers import SentenceTransformer
import faiss, numpy as np

embedder = SentenceTransformer('all-MiniLM-L6-v2')

def create_vector_db(documents):
    embeddings = embedder.encode(documents)
    index = faiss.IndexFlatL2(embeddings.shape[1])
    index.add(embeddings.astype(np.float32))
    return index, documents

import openai, faiss, numpy as np
from sentence_transformers import SentenceTransformer

client = openai.OpenAI(
    api_key="YOUR_AICREDITS_API_KEY",
    base_url="https://api.aicreditsapi.com/v1"
)
embedder = SentenceTransformer('all-MiniLM-L6-v2')

docs = ["Your documents here..."]
embeddings = embedder.encode(docs)
index = faiss.IndexFlatL2(embeddings.shape[1])
index.add(embeddings.astype(np.float32))

def rag(query):
    q_emb = embedder.encode([query])
    _, I = index.search(q_emb.astype(np.float32), k=3)
    context = "\n".join([docs[i] for i in I[0]])
    response = client.chat.completions.create(
        model="deepseek-chat",
        messages=[{"role": "user", "content": f"Context: {context}\n\nQuestion: {query}"}],
        max_tokens=500
    )
    return response.choices[0].message.content

print(rag("What is DeepSeek?"))