14 files changed, 219 insertions, 269 deletions

diff --git a/g4f/Provider/AItianhuSpace.py b/g4f/Provider/AItianhuSpace.py
index fabe6b47..95386e8e 100644
--- a/g4f/Provider/AItianhuSpace.py
+++ b/g4f/Provider/AItianhuSpace.py
@@ -5,7 +5,8 @@ import random
 
 from ..typing import CreateResult, Messages
 from .base_provider import BaseProvider
-from .helper import WebDriver, WebDriverSession, format_prompt, get_random_string
+from .helper import format_prompt, get_random_string
+from .webdriver import WebDriver, WebDriverSession
 from .. import debug
 
 class AItianhuSpace(BaseProvider):
@@ -24,7 +25,7 @@ class AItianhuSpace(BaseProvider):
         domain: str = None,
         proxy: str = None,
         timeout: int = 120,
-        web_driver: WebDriver = None,
+        webdriver: WebDriver = None,
         headless: bool = True,
         **kwargs
     ) -> CreateResult:
@@ -39,7 +40,7 @@ class AItianhuSpace(BaseProvider):
         url = f"https://{domain}"
         prompt = format_prompt(messages)
 
-        with WebDriverSession(web_driver, "", headless=headless, proxy=proxy) as driver:
+        with WebDriverSession(webdriver, "", headless=headless, proxy=proxy) as driver:
             from selenium.webdriver.common.by import By
             from selenium.webdriver.support.ui import WebDriverWait
             from selenium.webdriver.support import expected_conditions as EC
diff --git a/g4f/Provider/MyShell.py b/g4f/Provider/MyShell.py
index a1c8d335..5c9c4fe6 100644
--- a/g4f/Provider/MyShell.py
+++ b/g4f/Provider/MyShell.py
@@ -4,7 +4,8 @@ import time, json
 
 from ..typing import CreateResult, Messages
 from .base_provider import BaseProvider
-from .helper import WebDriver, WebDriverSession, format_prompt
+from .helper import format_prompt
+from .webdriver import WebDriver, WebDriverSession
 
 class MyShell(BaseProvider):
     url = "https://app.myshell.ai/chat"
@@ -20,10 +21,10 @@ class MyShell(BaseProvider):
         stream: bool,
         proxy: str = None,
         timeout: int = 120,
-        web_driver: WebDriver = None,
+        webdriver: WebDriver = None,
         **kwargs
     ) -> CreateResult:
-        with WebDriverSession(web_driver, "", proxy=proxy) as driver:
+        with WebDriverSession(webdriver, "", proxy=proxy) as driver:
             from selenium.webdriver.common.by import By
             from selenium.webdriver.support.ui import WebDriverWait
             from selenium.webdriver.support import expected_conditions as EC
@@ -52,15 +53,16 @@ response = await fetch("https://api.myshell.ai/v1/bot/chat/send_message", {
     "body": '{body}',
     "method": "POST"
 })
-window.reader = response.body.getReader();
+window._reader = response.body.pipeThrough(new TextDecoderStream()).getReader();
 """
             driver.execute_script(script.replace("{body}", json.dumps(data)))
             script = """
-chunk = await window.reader.read();
-if (chunk['done']) return null;
-text = (new TextDecoder()).decode(chunk['value']);
+chunk = await window._reader.read();
+if (chunk['done']) {
+    return null;
+}
 content = '';
-text.split('\\n').forEach((line, index) => {
+chunk['value'].split('\\n').forEach((line, index) => {
     if (line.startsWith('data: ')) {
         try {
             const data = JSON.parse(line.substring('data: '.length));
diff --git a/g4f/Provider/PerplexityAi.py b/g4f/Provider/PerplexityAi.py
index c0b2412e..03353a95 100644
--- a/g4f/Provider/PerplexityAi.py
+++ b/g4f/Provider/PerplexityAi.py
@@ -4,7 +4,8 @@ import time
 
 from ..typing import CreateResult, Messages
 from .base_provider import BaseProvider
-from .helper import WebDriver, WebDriverSession, format_prompt
+from .helper import format_prompt
+from .webdriver import WebDriver, WebDriverSession
 
 class PerplexityAi(BaseProvider):
     url = "https://www.perplexity.ai"
@@ -20,12 +21,12 @@ class PerplexityAi(BaseProvider):
         stream: bool,
         proxy: str = None,
         timeout: int = 120,
-        web_driver: WebDriver = None,
+        webdriver: WebDriver = None,
         virtual_display: bool = True,
         copilot: bool = False,
         **kwargs
     ) -> CreateResult:
-        with WebDriverSession(web_driver, "", virtual_display=virtual_display, proxy=proxy) as driver:
+        with WebDriverSession(webdriver, "", virtual_display=virtual_display, proxy=proxy) as driver:
             from selenium.webdriver.common.by import By
             from selenium.webdriver.support.ui import WebDriverWait
             from selenium.webdriver.support import expected_conditions as EC
diff --git a/g4f/Provider/Phind.py b/g4f/Provider/Phind.py
index 32f63665..82769ab0 100644
--- a/g4f/Provider/Phind.py
+++ b/g4f/Provider/Phind.py
@@ -5,7 +5,8 @@ from urllib.parse import quote
 
 from ..typing import CreateResult, Messages
 from .base_provider import BaseProvider
-from .helper import WebDriver, WebDriverSession, format_prompt
+from .helper import format_prompt
+from .webdriver import WebDriver, WebDriverSession
 
 class Phind(BaseProvider):
     url = "https://www.phind.com"
@@ -21,11 +22,11 @@ class Phind(BaseProvider):
         stream: bool,
         proxy: str = None,
         timeout: int = 120,
-        web_driver: WebDriver = None,
+        webdriver: WebDriver = None,
         creative_mode: bool = None,
         **kwargs
     ) -> CreateResult:
-        with WebDriverSession(web_driver, "", proxy=proxy) as driver:
+        with WebDriverSession(webdriver, "", proxy=proxy) as driver:
             from selenium.webdriver.common.by import By
             from selenium.webdriver.support.ui import WebDriverWait
             from selenium.webdriver.support import expected_conditions as EC
@@ -34,40 +35,38 @@ class Phind(BaseProvider):
             driver.get(f"{cls.url}/search?q={prompt}&source=searchbox")
 
             # Register fetch hook
-            driver.execute_script("""
+            source = """
 window._fetch = window.fetch;
-window.fetch = (url, options) => {
-    // Call parent fetch method
-    const result = window._fetch(url, options);
+window.fetch = async (url, options) => {
+    const response = await window._fetch(url, options);
     if (url != "/api/infer/answer") {
-        return result;
+        return response;
     }
-    // Load response reader
-    result.then((response) => {
-        if (!response.body.locked) {
-            window._reader = response.body.getReader();
-        }
-    });
-    // Return dummy response
-    return new Promise((resolve, reject) => {
-        resolve(new Response(new ReadableStream()))
-    });
+    copy = response.clone();
+    window._reader = response.body.pipeThrough(new TextDecoderStream()).getReader();
+    return copy;
 }
-""")
+"""
+            driver.execute_cdp_cmd("Page.addScriptToEvaluateOnNewDocument", {
+                "source": source
+            })
 
             # Need to change settings
-            if model.startswith("gpt-4") or creative_mode:
-                wait = WebDriverWait(driver, timeout)
+            wait = WebDriverWait(driver, timeout)
+            def open_dropdown():
                 # Open settings dropdown
                 wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, "button.text-dark.dropdown-toggle")))
                 driver.find_element(By.CSS_SELECTOR, "button.text-dark.dropdown-toggle").click()
                 # Wait for dropdown toggle
                 wait.until(EC.visibility_of_element_located((By.XPATH, "//button[text()='GPT-4']")))
-                # Enable GPT-4
+            if model.startswith("gpt-4") or creative_mode:
+               # Enable GPT-4
                 if model.startswith("gpt-4"):
+                    open_dropdown()
                     driver.find_element(By.XPATH, "//button[text()='GPT-4']").click()
                 # Enable creative mode
                 if creative_mode or creative_mode == None:
+                    open_dropdown()
                     driver.find_element(By.ID, "Creative Mode").click()
                 # Submit changes
                 driver.find_element(By.CSS_SELECTOR, ".search-bar-input-group button[type='submit']").click()
@@ -78,10 +77,11 @@ window.fetch = (url, options) => {
                 chunk = driver.execute_script("""
 if(window._reader) {
     chunk = await window._reader.read();
-    if (chunk['done']) return null;
-    text = (new TextDecoder()).decode(chunk['value']);
+    if (chunk['done']) {
+        return null;
+    }
     content = '';
-    text.split('\\r\\n').forEach((line, index) => {
+    chunk['value'].split('\\r\\n').forEach((line, index) => {
         if (line.startsWith('data: ')) {
             line = line.substring('data: '.length);
             if (!line.startsWith('<PHIND_METADATA>')) {
diff --git a/g4f/Provider/TalkAi.py b/g4f/Provider/TalkAi.py
index 20ba65b5..0edd9f6b 100644
--- a/g4f/Provider/TalkAi.py
+++ b/g4f/Provider/TalkAi.py
@@ -4,7 +4,7 @@ import time, json, time
 
 from ..typing import CreateResult, Messages
 from .base_provider import BaseProvider
-from .helper import WebDriver, WebDriverSession
+from .webdriver import WebDriver, WebDriverSession
 
 class TalkAi(BaseProvider):
     url = "https://talkai.info"
@@ -19,10 +19,10 @@ class TalkAi(BaseProvider):
         messages: Messages,
         stream: bool,
         proxy: str = None,
-        web_driver: WebDriver = None,
+        webdriver: WebDriver = None,
         **kwargs
     ) -> CreateResult:
-        with WebDriverSession(web_driver, "", virtual_display=True, proxy=proxy) as driver:
+        with WebDriverSession(webdriver, "", virtual_display=True, proxy=proxy) as driver:
             from selenium.webdriver.common.by import By
             from selenium.webdriver.support.ui import WebDriverWait
             from selenium.webdriver.support import expected_conditions as EC
diff --git a/g4f/Provider/helper.py b/g4f/Provider/helper.py
index 03e9ba94..2171f0b7 100644
--- a/g4f/Provider/helper.py
+++ b/g4f/Provider/helper.py
@@ -6,7 +6,6 @@ import webbrowser
 import random
 import string
 import secrets
-import time
 from os              import path
 from asyncio         import AbstractEventLoop
 from platformdirs    import user_config_dir
@@ -21,26 +20,8 @@ from browser_cookie3 import (
     firefox,
     BrowserCookieError
 )
-try: 
-    from selenium.webdriver.remote.webdriver import WebDriver 
-except ImportError: 
-    class WebDriver(): 
-        pass
-try:
-    from undetected_chromedriver import Chrome, ChromeOptions
-except ImportError:
-    class Chrome():
-        def __init__():
-            raise RuntimeError('Please install the "undetected_chromedriver" package')
-    class ChromeOptions():
-        def add_argument():
-            pass
-try:
-    from pyvirtualdisplay import Display
-except ImportError:
-    pass
 
-from ..typing import Dict, Messages, Union, Tuple
+from ..typing import Dict, Messages
 from .. import debug
 
 # Change event loop policy on windows
@@ -135,74 +116,11 @@ def format_prompt(messages: Messages, add_special_tokens=False) -> str:
     return f"{formatted}\nAssistant:"
 
 
-def get_browser(
-    user_data_dir: str = None,
-    headless: bool = False,
-    proxy: str = None,
-    options: ChromeOptions = None
-) -> Chrome:
-    if user_data_dir == None:
-        user_data_dir = user_config_dir("g4f")
-    if proxy:
-        if not options:
-            options = ChromeOptions()
-        options.add_argument(f'--proxy-server={proxy}')
-    return Chrome(options=options, user_data_dir=user_data_dir, headless=headless)
-
-class WebDriverSession():
-    def __init__(
-        self,
-        web_driver: WebDriver = None,
-        user_data_dir: str = None,
-        headless: bool = False,
-        virtual_display: bool = False,
-        proxy: str = None,
-        options: ChromeOptions = None
-    ):
-        self.web_driver = web_driver
-        self.user_data_dir = user_data_dir
-        self.headless = headless
-        self.virtual_display = virtual_display
-        self.proxy = proxy
-        self.options = options
-    
-    def reopen(
-        self,
-        user_data_dir: str = None,
-        headless: bool = False,
-        virtual_display: bool = False
-    ) -> WebDriver:
-        if user_data_dir == None:
-            user_data_dir = self.user_data_dir
-        self.default_driver.quit()
-        if not virtual_display and self.virtual_display:
-            self.virtual_display.stop()
-        self.default_driver = get_browser(user_data_dir, headless, self.proxy)
-        return self.default_driver
-
-    def __enter__(self) -> WebDriver:
-        if self.web_driver:
-            return self.web_driver
-        if self.virtual_display == True:
-            self.virtual_display = Display(size=(1920,1080))
-            self.virtual_display.start()
-        self.default_driver = get_browser(self.user_data_dir, self.headless, self.proxy, self.options)
-        return self.default_driver
-
-    def __exit__(self, exc_type, exc_val, exc_tb):
-        if self.default_driver:
-            self.default_driver.close()
-            time.sleep(0.1)
-            self.default_driver.quit()
-        if self.virtual_display:
-            self.virtual_display.stop()
-
 def get_random_string(length: int = 10) -> str:
     return ''.join(
         random.choice(string.ascii_lowercase + string.digits)
         for _ in range(length)
     )
 
-
 def get_random_hex() -> str:
     return secrets.token_hex(16).zfill(32)
\ No newline at end of file
diff --git a/g4f/Provider/needs_auth/Bard.py b/g4f/Provider/needs_auth/Bard.py
index 77c029b8..2c1f6121 100644
--- a/g4f/Provider/needs_auth/Bard.py
+++ b/g4f/Provider/needs_auth/Bard.py
@@ -4,7 +4,8 @@ import time
 
 from ...typing import CreateResult, Messages
 from ..base_provider import BaseProvider
-from ..helper import WebDriver, WebDriverSession, format_prompt
+from ..helper import format_prompt
+from ..webdriver import WebDriver, WebDriverSession
 
 class Bard(BaseProvider):
     url = "https://bard.google.com"
@@ -18,13 +19,13 @@ class Bard(BaseProvider):
         messages: Messages,
         stream: bool,
         proxy: str = None,
-        web_driver: WebDriver = None,
+        webdriver: WebDriver = None,
         user_data_dir: str = None,
         headless: bool = True,
         **kwargs
     ) -> CreateResult:
         prompt = format_prompt(messages)
-        session = WebDriverSession(web_driver, user_data_dir, headless, proxy=proxy)
+        session = WebDriverSession(webdriver, user_data_dir, headless, proxy=proxy)
         with session as driver:
             from selenium.webdriver.common.by import By
             from selenium.webdriver.support.ui import WebDriverWait
@@ -36,8 +37,8 @@ class Bard(BaseProvider):
                 wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, "div.ql-editor.textarea")))
             except:
                 # Reopen browser for login
-                if not web_driver:
-                    driver = session.reopen(headless=False)
+                if not webdriver:
+                    driver = session.reopen()
                     driver.get(f"{cls.url}/chat")
                     wait = WebDriverWait(driver, 240)
                     wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, "div.ql-editor.textarea")))
diff --git a/g4f/Provider/needs_auth/HuggingChat.py b/g4f/Provider/needs_auth/HuggingChat.py
index 68c6713b..59e2da73 100644
--- a/g4f/Provider/needs_auth/HuggingChat.py
+++ b/g4f/Provider/needs_auth/HuggingChat.py
@@ -59,17 +59,4 @@ class HuggingChat(AsyncGeneratorProvider):
                         break
                 
             async with session.delete(f"{cls.url}/conversation/{conversation_id}", proxy=proxy) as response:
-                response.raise_for_status()
-
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("proxy", "str"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                response.raise_for_status()
\ No newline at end of file
diff --git a/g4f/Provider/needs_auth/OpenAssistant.py b/g4f/Provider/needs_auth/OpenAssistant.py
index de62636c..e549b517 100644
--- a/g4f/Provider/needs_auth/OpenAssistant.py
+++ b/g4f/Provider/needs_auth/OpenAssistant.py
@@ -87,15 +87,3 @@ class OpenAssistant(AsyncGeneratorProvider):
             }
             async with session.delete("https://open-assistant.io/api/chat", proxy=proxy, params=params) as response:
                 response.raise_for_status()
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("proxy", "str"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
diff --git a/g4f/Provider/needs_auth/OpenaiChat.py b/g4f/Provider/needs_auth/OpenaiChat.py
index 9fd90812..8c9dd1e0 100644
--- a/g4f/Provider/needs_auth/OpenaiChat.py
+++ b/g4f/Provider/needs_auth/OpenaiChat.py
@@ -6,7 +6,8 @@ from asyncstdlib.itertools import tee
 from async_property import async_cached_property
 
 from ..base_provider import AsyncGeneratorProvider
-from ..helper import get_browser, get_event_loop
+from ..helper import get_event_loop
+from ..webdriver import get_browser
 from ...typing import AsyncResult, Messages
 from ...requests import StreamSession
 
@@ -38,7 +39,10 @@ class OpenaiChat(AsyncGeneratorProvider):
         **kwargs
     ) -> Response:
         if prompt:
-            messages.append({"role": "user", "content": prompt})
+            messages.append({
+                "role": "user",
+                "content": prompt
+            })
         generator = cls.create_async_generator(
             model,
             messages,
@@ -49,12 +53,9 @@ class OpenaiChat(AsyncGeneratorProvider):
             response_fields=True,
             **kwargs
         )
-        fields: ResponseFields = await anext(generator)
-        if "access_token" not in kwargs:
-            kwargs["access_token"] = cls._access_token
         return Response(
             generator,
-            fields,
+            await anext(generator),
             action,
             messages,
             kwargs
@@ -87,7 +88,6 @@ class OpenaiChat(AsyncGeneratorProvider):
         headers = {
             "Accept": "text/event-stream",
             "Authorization": f"Bearer {access_token}",
-            "Cookie": 'intercom-device-id-dgkjq2bp=0f047573-a750-46c8-be62-6d54b56e7bf0; ajs_user_id=user-iv3vxisaoNodwWpxmNpMfekH; ajs_anonymous_id=fd91be0b-0251-4222-ac1e-84b1071e9ec1; __Host-next-auth.csrf-token=d2b5f67d56f7dd6a0a42ae4becf2d1a6577b820a5edc88ab2018a59b9b506886%7Ce5c33eecc460988a137cbc72d90ee18f1b4e2f672104f368046df58e364376ac; _cfuvid=gt_mA.q6rue1.7d2.AR0KHpbVBS98i_ppfi.amj2._o-1700353424353-0-604800000; cf_clearance=GkHCfPSFU.NXGcHROoe4FantnqmnNcluhTNHz13Tk.M-1700353425-0-1-dfe77f81.816e9bc2.714615da-0.2.1700353425; __Secure-next-auth.callback-url=https%3A%2F%2Fchat.openai.com; intercom-session-dgkjq2bp=UWdrS1hHazk5VXN1c0V5Q1F0VXdCQmsyTU9pVjJMUkNpWnFnU3dKWmtIdGwxTC9wbjZuMk5hcEc0NWZDOGdndS0tSDNiaDNmMEdIL1RHU1dFWDBwOHFJUT09--f754361b91fddcd23a13b288dcb2bf8c7f509e91; _uasid="Z0FBQUFBQmxXVnV0a3dmVno4czRhcDc2ZVcwaUpSNUdZejlDR25YSk5NYTJQQkpyNmRvOGxjTHMyTlAxWmJhaURrMVhjLXZxQXdZeVpBbU1aczA5WUpHT2dwaS1MOWc4MnhyNWFnbGRzeGdJcGFKT0ZRdnBTMVJHcGV2MGNTSnVQY193c0hqUWIycHhQRVF4dENlZ3phcDdZeHgxdVhoalhrZmtZME9NbWhMQjdVR3Vzc3FRRk0ybjJjNWMwTWtIRjdPb19lUkFtRmV2MDVqd1kwWU11QTYtQkdZenEzVHhLMGplY1hZM3FlYUt1cVZaNWFTRldleEJETzJKQjk1VTJScy1GUnMxUVZWMnVxYklxMjdockVZbkZyd1R4U1RtMnA1ZzlSeXphdmVOVk9xeEdrRkVOSjhwTVd1QzFtQjhBcWdDaE92Q1VlM2pwcjFQTXRuLVJNRVlZSGpIdlZ0aGV3PT0="; _dd_s=rum=0&expire=1700356244884; __Secure-next-auth.session-token=eyJhbGciOiJkaXIiLCJlbmMiOiJBMjU2R0NNIn0..3aK6Fbdy2_8f07bf.8eT2xgonrCnz7ySY6qXFsg3kzL6UQfXKAYaw3tyn-6_X9657zy47k9qGvmi9mF0QKozj5jau3_Ca62AQQ7FmeC6Y2F1urtzqrXqwTTsQ2LuzFPIQkx6KKb2DXc8zW2-oyEzJ_EY5yxfLB2RlRkSh3M7bYNZh4_ltEcfkj38s_kIPGMxv34udtPWGWET99MCjkdwQWXylJag4s0fETA0orsBAKnGCyqAUNJbb_D7BYtGSV-MQ925kZMG6Di_QmfO0HQWURDYjmdRNcuy1PT_xJ1DJko8sjL42i4j3RhkNDkhqCIqyYImz2eHFWHW7rYKxTkrBhlCPMS5hRdcCswD7JYPcSBiwnVRYgyOocFGXoFvQgIZ2FX9NiZ3SMEVM1VwIGSE-qH0H2nMa8_iBvsOgOWJgKjVAvzzyzZvRVDUUHzJrikSFPNONVDU3h-04c1kVL4qIu9DfeTPN7n8AvNmYwMbro0L9-IUAeXNo4-pwF0Kt-AtTsamqWvMqnK4O_YOyLnDDlvkmnOvDC2d5uinwlQIxr6APO6qFfGLlHiLZemKoekxEE1Fx70dl-Ouhk1VIzbF3OC6XNNxeBm9BUYUiHdL0wj2H9rHgX4cz6ZmS_3VTgpD6UJh-evu5KJ2gIvjYmVbyzEN0aPNDxfvBaOm-Ezpy4bUJ2bUrOwNn-0knWkDiTvjYmNhCyefPCtCF6rpKNay8PCw_yh79C4SdEP6Q4V7LI0Tvdi5uz7kLCiBC4AT9L0ao1WDX03mkUOpjvzHDvPLmj8chW3lTVm_kA0eYGQY4wT0jzleWlfV0Q8rB2oYECNLWksA3F1zlGfcl4lQjprvTXRePkvAbMpoJEsZD3Ylq7-foLDLk4-M2LYAFZDs282AY04sFjAjQBxTELFCCuDgTIgTXSIskY_XCxpVXDbdLlbCJY7XVK45ybwtfqwlKRp8Mo0B131uQAFc-migHaUaoGujxJJk21bP8F0OmhNYHBo4FQqE1rQm2JH5bNM7txKeh5KXdJgVUVbRSr7OIp_OF5-Bx_v9eRBGAIDkue26E2-O8Rnrp5zQ5TnvecQLDaUzWavCLPwsZ0_gsOLBxNOmauNYZtF8IElCsQSFDdhoiMxXsYUm4ZYKEAy3GWq8HGTAvBhNkh1hvnI7y-d8-DOaZf_D_D98-olZfm-LUkeosLNpPB9rxYMqViCiW3KrXE9Yx0wlFm5ePKaVvR7Ym_EPhSOhJBKFPCvdTdMZSNPUcW0ZJBVByq0A9sxD51lYq3gaFyqh94S4s_ox182AQ3szGzHkdgLcnQmJG9OYvKxAVcd43eg6_gODAYhx02GjbMw-7JTAhyXSeCrlMteHyOXl8hai-3LilC3PmMzi7Vbu49dhF1s4LcVlUowen5ira44rQQaB26mdaOUoQfodgt66M3RTWGPXyK1Nb72AzSXsCKyaQPbzeb6cN0fdGSdG4ktwvR04eFNEkquo_3aKu2GmUKTD0XcRx9dYrfXjgY-X1DDTVs1YND2gRhdx7FFEeBVjtbj2UqmG3Rvd4IcHGe7OnYWw2MHDcol68SsR1KckXWwWREz7YTGUnDB2M1kx_H4W2mjclytnlHOnYU3RflegRPeSTbdzUZJvGKXCCz45luHkQWN_4DExE76D-9YqbFIz-RY5yL4h-Zs-i2xjm2K-4xCMM9nQIOqhLMqixIZQ2ldDAidKoYtbs5ppzbcBLyrZM96bq9DwRBY3aacqWdlRd-TfX0wv5KO4fo0sSh5FsuhuN0zcEV_NNXgqIEM_p14EcPqgbrAvCBQ8os70TRBQLXiF0EniSofGjxwF8kQvUk3C6Wfc8cTTeN-E6GxCVTn91HBwA1iSEZlRLMVb8_BcRJNqwbgnb_07jR6-eo42u88CR3KQdAWwbQRdMxsURFwZ0ujHXVGG0Ll6qCFBcHXWyDO1x1yHdHnw8_8yF26pnA2iPzrFR-8glMgIA-639sLuGAxjO1_ZuvJ9CAB41Az9S_jaZwaWy215Hk4-BRYD-MKmHtonwo3rrxhE67WJgbbu14efsw5nT6ow961pffgwXov5VA1Rg7nv1E8RvQOx7umWW6o8R4W6L8f2COsmPTXfgwIjoJKkjhUqAQ8ceG7cM0ET-38yaC0ObU8EkXfdGGgxI28qTEZWczG66_iM4hw7QEGCY5Cz2kbO6LETAiw9OsSigtBvDS7f0Ou0bZ41pdK7G3FmvdZAnjWPjObnDF4k4uWfn7mzt0fgj3FyqK20JezRDyGuAbUUhOvtZpc9sJpzxR34eXEZTouuALrHcGuNij4z6rx51FrQsaMtiup8QVrhtZbXtKLMYnWYSbkhuTeN2wY-xV1ZUsQlakIZszzGF7kuIG87KKWMpuPMvbXjz6Pp_gWJiIC6aQuk8xl5g0iBPycf_6Q-MtpuYxzNE2TpI1RyR9mHeXmteoRzrFiWp7yEC-QGNFyAJgxTqxM3CjHh1Jt6IddOsmn89rUo1dZM2Smijv_fbIv3avXLkIPX1KZjILeJCtpU0wAdsihDaRiRgDdx8fG__F8zuP0n7ziHas73cwrfg-Ujr6DhC0gTNxyd9dDA_oho9N7CQcy6EFmfNF2te7zpLony0859jtRv2t1TnpzAa1VvMK4u6mXuJ2XDo04_6GzLO3aPHinMdl1BcIAWnqAqWAu3euGFLTHOhXlfijut9N1OCifd_zWjhVtzlR39uFeCQBU5DyQArzQurdoMx8U1ETsnWgElxGSStRW-YQoPsAJ87eg9trqKspFpTVlAVN3t1GtoEAEhcwhe81SDssLmKGLc.7PqS6jRGTIfgTPlO7Ognvg; __cf_bm=VMWoAKEB45hQSwxXtnYXcurPaGZDJS4dMi6dIMFLwdw-1700355394-0-ATVsbq97iCaTaJbtYr8vtg1Zlbs3nLrJLKVBHYa2Jn7hhkGclqAy8Gbyn5ePEhDRqj93MsQmtayfYLqY5n4WiLY=; __cflb=0H28vVfF4aAyg2hkHFH9CkdHRXPsfCUf6VpYf2kz3RX'
         }
         async with StreamSession(
             proxies={"https": proxy},
@@ -95,24 +95,22 @@ class OpenaiChat(AsyncGeneratorProvider):
             headers=headers,
             timeout=timeout
         ) as session:
-            data = {
-                "action": action,
-                "arkose_token": await get_arkose_token(proxy, timeout),
-                "conversation_id": conversation_id,
-                "parent_message_id": parent_id,
-                "model": models[model],
-                "history_and_training_disabled": history_disabled and not auto_continue,
-            }
-            if action != "continue":
-                data["messages"] = [{
-                    "id": str(uuid.uuid4()),
-                    "author": {"role": "user"},
-                    "content": {"content_type": "text", "parts": [messages[-1]["content"]]},
-                }]
-            first = True
             end_turn = EndTurn()
-            while first or auto_continue and not end_turn.is_end:
-                first = False
+            while not end_turn.is_end:
+                data = {
+                    "action": action,
+                    "arkose_token": await get_arkose_token(proxy, timeout),
+                    "conversation_id": conversation_id,
+                    "parent_message_id": parent_id,
+                    "model": models[model],
+                    "history_and_training_disabled": history_disabled and not auto_continue,
+                }
+                if action != "continue":
+                    data["messages"] = [{
+                        "id": str(uuid.uuid4()),
+                        "author": {"role": "user"},
+                        "content": {"content_type": "text", "parts": [messages[-1]["content"]]},
+                    }]
                 async with session.post(f"{cls.url}/backend-api/conversation", json=data) as response:
                     try:
                         response.raise_for_status()
@@ -120,43 +118,38 @@ class OpenaiChat(AsyncGeneratorProvider):
                         raise RuntimeError(f"Error {response.status_code}: {await response.text()}")
                     last_message = 0
                     async for line in response.iter_lines():
-                        if line.startswith(b"data: "):
-                            line = line[6:]
-                            if line == b"[DONE]":
-                                break
-                            try:
-                                line = json.loads(line)
-                            except:
-                                continue
-                            if "message" not in line:
-                                continue
-                            if "error" in line and line["error"]:
-                                raise RuntimeError(line["error"])
-                            if "message_type" not in line["message"]["metadata"]:
-                                continue
-                            if line["message"]["author"]["role"] != "assistant":
-                                continue
-                            if line["message"]["metadata"]["message_type"] in ("next", "continue", "variant"):
-                                conversation_id = line["conversation_id"]
-                                parent_id = line["message"]["id"]
-                                if response_fields:
-                                    response_fields = False
-                                    yield ResponseFields(conversation_id, parent_id, end_turn)
-                                new_message = line["message"]["content"]["parts"][0]
-                                yield new_message[last_message:]
-                                last_message = len(new_message)
-                            if "finish_details" in line["message"]["metadata"]:
-                                if line["message"]["metadata"]["finish_details"]["type"] == "max_tokens":
-                                    end_turn.end()
-
-                data = {
-                    "action": "continue",
-                    "arkose_token": await get_arkose_token(proxy, timeout),
-                    "conversation_id": conversation_id,
-                    "parent_message_id": parent_id,
-                    "model": models[model],
-                    "history_and_training_disabled": False,
-                }
+                        if not line.startswith(b"data: "):
+                            continue
+                        line = line[6:]
+                        if line == b"[DONE]":
+                            break
+                        try:
+                            line = json.loads(line)
+                        except:
+                            continue
+                        if "message" not in line:
+                            continue
+                        if "error" in line and line["error"]:
+                            raise RuntimeError(line["error"])
+                        if "message_type" not in line["message"]["metadata"]:
+                            continue
+                        if line["message"]["author"]["role"] != "assistant":
+                            continue
+                        if line["message"]["metadata"]["message_type"] in ("next", "continue", "variant"):
+                            conversation_id = line["conversation_id"]
+                            parent_id = line["message"]["id"]
+                            if response_fields:
+                                response_fields = False
+                                yield ResponseFields(conversation_id, parent_id, end_turn)
+                            new_message = line["message"]["content"]["parts"][0]
+                            yield new_message[last_message:]
+                            last_message = len(new_message)
+                        if "finish_details" in line["message"]["metadata"]:
+                            if line["message"]["metadata"]["finish_details"]["type"] == "stop":
+                                end_turn.end()
+                if not auto_continue:
+                    break
+                action = "continue"
                 await asyncio.sleep(5)
 
     @classmethod
@@ -167,7 +160,7 @@ class OpenaiChat(AsyncGeneratorProvider):
                 from selenium.webdriver.support.ui import WebDriverWait
                 from selenium.webdriver.support import expected_conditions as EC
 
-                driver = get_browser("~/openai", proxy=proxy)
+                driver = get_browser(proxy=proxy)
             except ImportError:
                 return
             try:
@@ -193,18 +186,6 @@ class OpenaiChat(AsyncGeneratorProvider):
             raise RuntimeError("Read access token failed")
         return cls._access_token
 
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("proxy", "str"),
-            ("access_token", "str"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
     
 async def get_arkose_token(proxy: str = None, timeout: int = None) -> str:
     config = {
@@ -293,7 +274,7 @@ class Response():
     
     async def variant(self, **kwargs) -> Response:
         if self.action != "next":
-            raise RuntimeError("Can't create variant with continue or variant request.")
+            raise RuntimeError("Can't create variant from continue or variant request.")
         return await OpenaiChat.create(
             **self._options,
             messages=self._messages,
diff --git a/g4f/Provider/needs_auth/Poe.py b/g4f/Provider/needs_auth/Poe.py
index 1c8c97d7..99f6945b 100644
--- a/g4f/Provider/needs_auth/Poe.py
+++ b/g4f/Provider/needs_auth/Poe.py
@@ -4,7 +4,8 @@ import time
 
 from ...typing import CreateResult, Messages
 from ..base_provider import BaseProvider
-from ..helper import WebDriver, WebDriverSession, format_prompt
+from ..helper import format_prompt
+from ..webdriver import WebDriver, WebDriverSession
 
 models = {
     "meta-llama/Llama-2-7b-chat-hf": {"name": "Llama-2-7b"},
@@ -33,7 +34,7 @@ class Poe(BaseProvider):
         messages: Messages,
         stream: bool,
         proxy: str = None,
-        web_driver: WebDriver = None,
+        webdriver: WebDriver = None,
         user_data_dir: str = None,
         headless: bool = True,
         **kwargs
@@ -44,7 +45,7 @@ class Poe(BaseProvider):
             raise ValueError(f"Model are not supported: {model}")
         prompt = format_prompt(messages)
 
-        session = WebDriverSession(web_driver, user_data_dir, headless, proxy=proxy)
+        session = WebDriverSession(webdriver, user_data_dir, headless, proxy=proxy)
         with session as driver:
             from selenium.webdriver.common.by import By
             from selenium.webdriver.support.ui import WebDriverWait
@@ -80,8 +81,8 @@ class Poe(BaseProvider):
                 wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, "textarea[class^='GrowingTextArea']")))
             except:
                 # Reopen browser for login
-                if not web_driver:
-                    driver = session.reopen(headless=False)
+                if not webdriver:
+                    driver = session.reopen()
                     driver.get(f"{cls.url}/{models[model]['name']}")
                     wait = WebDriverWait(driver, 240)
                     wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, "textarea[class^='GrowingTextArea']")))
diff --git a/g4f/Provider/needs_auth/Raycast.py b/g4f/Provider/needs_auth/Raycast.py
index 4570fd9f..d7be98ac 100644
--- a/g4f/Provider/needs_auth/Raycast.py
+++ b/g4f/Provider/needs_auth/Raycast.py
@@ -60,18 +60,3 @@ class Raycast(BaseProvider):
             token = completion_chunk['text']
             if token != None:
                 yield token
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-            ("top_p", "int"),
-            ("model", "str"),
-            ("auth", "str"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
diff --git a/g4f/Provider/needs_auth/Theb.py b/g4f/Provider/needs_auth/Theb.py
index cf33f0c6..49ee174b 100644
--- a/g4f/Provider/needs_auth/Theb.py
+++ b/g4f/Provider/needs_auth/Theb.py
@@ -4,7 +4,8 @@ import time
 
 from ...typing import CreateResult, Messages
 from ..base_provider import BaseProvider
-from ..helper import WebDriver, WebDriverSession, format_prompt
+from ..helper import format_prompt
+from ..webdriver import WebDriver, WebDriverSession
 
 models = {
     "theb-ai": "TheB.AI",
@@ -44,14 +45,14 @@ class Theb(BaseProvider):
         messages: Messages,
         stream: bool,
         proxy: str = None,
-        web_driver: WebDriver = None,
+        webdriver: WebDriver = None,
         virtual_display: bool = True,
         **kwargs
     ) -> CreateResult:
         if model in models:
             model = models[model]
         prompt = format_prompt(messages)
-        web_session = WebDriverSession(web_driver, virtual_display=virtual_display, proxy=proxy)
+        web_session = WebDriverSession(webdriver, virtual_display=virtual_display, proxy=proxy)
         with web_session as driver:
             from selenium.webdriver.common.by import By
             from selenium.webdriver.support.ui import WebDriverWait
@@ -61,22 +62,16 @@ class Theb(BaseProvider):
             # Register fetch hook
             script = """
 window._fetch = window.fetch;
-window.fetch = (url, options) => {
+window.fetch = async (url, options) => {
     // Call parent fetch method
-    const result = window._fetch(url, options);
+    const response = await window._fetch(url, options);
     if (!url.startsWith("/api/conversation")) {
         return result;
     }
-    // Load response reader
-    result.then((response) => {
-        if (!response.body.locked) {
-            window._reader = response.body.getReader();
-        }
-    });
-    // Return dummy response
-    return new Promise((resolve, reject) => {
-        resolve(new Response(new ReadableStream()))
-    });
+    // Copy response
+    copy = response.clone();
+    window._reader = response.body.pipeThrough(new TextDecoderStream()).getReader();
+    return copy;
 }
 window._last_message = "";
 """
@@ -97,7 +92,6 @@ window._last_message = "";
                 wait = WebDriverWait(driver, 240)
                 wait.until(EC.visibility_of_element_located((By.ID, "textareaAutosize")))
 
-            time.sleep(200)
             try:
                 driver.find_element(By.CSS_SELECTOR, ".driver-overlay").click()
                 driver.find_element(By.CSS_SELECTOR, ".driver-overlay").click()
@@ -134,9 +128,8 @@ if(window._reader) {
     if (chunk['done']) {
         return null;
     }
-    text = (new TextDecoder()).decode(chunk['value']);
     message = '';
-    text.split('\\r\\n').forEach((line, index) => {
+    chunk['value'].split('\\r\\n').forEach((line, index) => {
         if (line.startsWith('data: ')) {
             try {
                 line = JSON.parse(line.substring('data: '.length));
diff --git a/g4f/Provider/webdriver.py b/g4f/Provider/webdriver.py
new file mode 100644
index 00000000..da3b13ed
--- /dev/null
+++ b/g4f/Provider/webdriver.py
@@ -0,0 +1,92 @@
+from __future__ import annotations
+
+import time
+from platformdirs import user_config_dir
+try: 
+    from selenium.webdriver.remote.webdriver import WebDriver 
+except ImportError: 
+    class WebDriver(): 
+        pass
+try:
+    from undetected_chromedriver import Chrome, ChromeOptions
+except ImportError:
+    class Chrome():
+        def __init__():
+            raise RuntimeError('Please install the "undetected_chromedriver" package')
+    class ChromeOptions():
+        def add_argument():
+            pass
+try:
+    from pyvirtualdisplay import Display
+    has_pyvirtualdisplay = True
+except ImportError:
+    has_pyvirtualdisplay = False
+
+def get_browser(
+    user_data_dir: str = None,
+    headless: bool = False,
+    proxy: str = None,
+    options: ChromeOptions = None
+) -> Chrome:
+    if user_data_dir == None:
+        user_data_dir = user_config_dir("g4f")
+    if proxy:
+        if not options:
+            options = ChromeOptions()
+        options.add_argument(f'--proxy-server={proxy}')
+    return Chrome(options=options, user_data_dir=user_data_dir, headless=headless)
+
+class WebDriverSession():
+    def __init__(
+        self,
+        webdriver: WebDriver = None,
+        user_data_dir: str = None,
+        headless: bool = False,
+        virtual_display: bool = False,
+        proxy: str = None,
+        options: ChromeOptions = None
+    ):
+        self.webdriver = webdriver
+        self.user_data_dir = user_data_dir
+        self.headless = headless
+        self.virtual_display = None
+        if has_pyvirtualdisplay and virtual_display:
+            self.virtual_display = Display(size=(1920,1080))
+        self.proxy = proxy
+        self.options = options
+        self.default_driver = None
+    
+    def reopen(
+        self,
+        user_data_dir: str = None,
+        headless: bool = False,
+        virtual_display: bool = False
+    ) -> WebDriver:
+        if user_data_dir == None:
+            user_data_dir = self.user_data_dir
+        if self.default_driver:
+            self.default_driver.quit()
+        if not virtual_display and self.virtual_display:
+            self.virtual_display.stop()
+            self.virtual_display = None
+        self.default_driver = get_browser(user_data_dir, headless, self.proxy)
+        return self.default_driver
+
+    def __enter__(self) -> WebDriver:
+        if self.webdriver:
+            return self.webdriver
+        if self.virtual_display:
+            self.virtual_display.start()
+        self.default_driver = get_browser(self.user_data_dir, self.headless, self.proxy, self.options)
+        return self.default_driver
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        if self.default_driver:
+            try:
+                self.default_driver.close()
+            except:
+                pass
+            time.sleep(0.1)
+            self.default_driver.quit()
+        if self.virtual_display:
+            self.virtual_display.stop()
\ No newline at end of file