diff --git a/package.json b/package.json
index 01a05ce..a105ea5 100644
--- a/package.json
+++ b/package.json
@@ -11,6 +11,7 @@
     "lint": "next lint"
   },
   "dependencies": {
+    "@google/generative-ai": "^0.21.0",
     "axios": "^1.7.9",
     "next": "15.1.0",
     "react": "^19.0.0",
diff --git a/src/llm/llm-config.js b/src/llm/llm-config.js
new file mode 100644
index 0000000..cfecd57
--- /dev/null
+++ b/src/llm/llm-config.js
@@ -0,0 +1,50 @@
+/**
+ * Configuration class for LLM generation parameters
+ * @class LLMConfig
+ * @property {number} temperature - Controls randomness in the output (0.0 to 2.0)
+ * @property {number} topP - Controls diversity via nucleus sampling (0.0 to 1.0)
+ * @property {number} topK - Controls diversity by limiting to K most likely tokens
+ * @property {number} maxToken - Controls the max number of output token
+ */
+export default class LLMConfig {
+    constructor(temperature, topP, topK, maxToken) {
+        if (temperature > 2 || temperature < 0) throw new Error('Temperature must be between 0.0 and 2.0');
+        if (topP > 1 || topP < 0) throw new Error('Top-p must be between 0.0 and 1.0');
+        this._temperature = temperature;
+        this._topP = topP;
+        this._topK = topK;
+        this._maxToken = maxToken;
+    }
+
+    /**
+     * Gets the temperature parameter
+     * @returns {number} The temperature value (0.0 to 1.0)
+     */
+    get temperature() {
+        return this._temperature;
+    }
+
+    /**
+     * Gets the top-p parameter
+     * @returns {number} The top-p value (0.0 to 1.0)
+     */
+    get topP() {
+        return this._topP;
+    }
+
+    /**
+     * Gets the top-k parameter
+     * @returns {number} The top-k value
+     */
+    get topK() {
+        return this._topK;
+    }
+
+    /**
+     * Gets the max output token parameter
+     * @returns {number} The number of tokens
+     */
+    get maxToken() {
+        return this._maxToken;
+    }
+}
\ No newline at end of file
diff --git a/src/llm/llm-provider.js b/src/llm/llm-provider.js
new file mode 100644
index 0000000..46d6ad8
--- /dev/null
+++ b/src/llm/llm-provider.js
@@ -0,0 +1,34 @@
+import LLMConfig from '@/llm/llmconfig.js'
+
+/**
+ * Abstract class for LLM providers (e.g. OpenAI, Gemini, etc.)
+ * @abstract
+ * @class LLMProvider
+ */
+export default class LLMProvider {
+    /**
+     * Constructor for LLMProvider
+     * @param {string} apikey - API key for the LLM service (your secret key)
+     * @param {string} model - Model identifier (e.g. gemini-1.5-pro)
+     * @param {LLMConfig} llmconfig - Configuration for LLM
+     * @param {string} systemPrompt - System prompt for the LLM
+     */
+    constructor(apikey, model, llmconfig, systemPrompt) {
+        this.apikey = apikey;
+        this.model = model;
+        this.llmconfig = llmconfig;
+        this.systemPrompt = systemPrompt
+    }
+
+    /**
+     * Executes the LLM with given prompt
+     * @abstract
+     * @param {string} userPrompt - User input prompt (The code will go inside here)
+     * @param {Array<role: string, parts: Array<text: string>>} history - The history of the conversation
+     * @throws {Error} When not implemented by child class
+     * @returns {Promise<string>} The LLM response
+     */
+    async run(userPrompt, history) {
+        throw new Error('The LLM chat method run() must be implemented');
+    }
+}
\ No newline at end of file
diff --git a/src/llm/provider/google.js b/src/llm/provider/google.js
new file mode 100644
index 0000000..e5142c1
--- /dev/null
+++ b/src/llm/provider/google.js
@@ -0,0 +1,43 @@
+import LLMProvider from '@/llm/llm-provider.js';
+import LLMConfig from '@/llm/llm-config.js';
+import { GoogleGenerativeAI } from '@google/generative-ai';
+
+/**
+ * Class for Google Gemini LLM
+ * @class GoogleProvider
+ */
+class GoogleProvider extends LLMProvider {
+    /**
+     * Constructor for Google Gemini LLM
+     * @param {string} apiKey - API key for google ai studio (https://aistudio.google.com)
+     * @param {string} modelName - Select model available in google ai studio
+     * @param {LLMConfig} llmconfig - Configuration for LLM
+     * @param {string} systemPrompt - System prompt for the LLM
+     */
+    constructor(apiKey, modelName, llmconfig, systemPrompt) {
+        super(apiKey, modelName, llmconfig, systemPrompt);
+        this.llm = new GoogleGenerativeAI({ apiKey: apiKey, systemInstruction: systemPrompt, model: modelName });
+
+        this.config = {
+            temperature: llmconfig.temperature,
+            topP: llmconfig.topP,
+            topK: llmconfig.topK,
+            maxOutputTokens: llmconfig.maxOutputTokens,
+            responseMimeType: "text/plain",
+        }
+    }
+
+    /**
+     * Executes the LLM with given prompt
+     * @param {string} userPrompt - User input prompt (The code will go inside here)
+     * @param {Array<role: string, parts: Array<text: string>>} history - The history of the conversation
+     * @returns {Promise<string>} The LLM response
+     */
+    async run(userPrompt, history) {
+        const chatSession = this.llm.startChat({
+           ...this.config,
+           history: history
+        });
+        return await chatSession.sendMessage(userPrompt);
+    }
+}
diff --git a/src/test/llm/llm-config.test.js b/src/test/llm/llm-config.test.js
new file mode 100644
index 0000000..463034f
--- /dev/null
+++ b/src/test/llm/llm-config.test.js
@@ -0,0 +1,21 @@
+import LLMConfig from '@/llm/llm-config.js';
+
+describe('LLMConfig', () => {
+    describe('constructor validation', () => {
+        it('should create instance with valid parameters', () => {
+            const config = new LLMConfig(0.7, 0.9, 50, 1000);
+            expect(config).toBeInstanceOf(LLMConfig);
+        });
+
+        it('should throw error for temperature < 0', () => {
+            expect(() => new LLMConfig(-0.1, 0.9, 50, 1000))
+                .toThrow('Temperature must be between 0.0 and 2.0');
+        });
+
+        it('should throw error for temperature > 2', () => {
+            expect(() => new LLMConfig(2.1, 0.9, 50, 1000))
+                .toThrow('Temperature must be between 0.0 and 2.0');
+        });
+
+    });
+});
\ No newline at end of file