Merge pull request #1274 from hlohaus/webdriver

Webdriver module, translate readme, support stream in create_async

README-DE.md

@@ -0,0 +1,726 @@
+<a href="./README.md">
+    <img src="https://img.shields.io/badge/open in-🇬🇧 english-blue.svg" alt="Open in EN">
+</a>
+
+![248433934-7886223b-c1d1-4260-82aa-da5741f303bb](https://github.com/xtekky/gpt4free/assets/98614666/ea012c87-76e0-496a-8ac4-e2de090cc6c9)
+
+<a href='https://ko-fi.com/xtekky' target='_blank'><img height='35' style='border:0px;height:46px;' src='https://az743702.vo.msecnd.net/cdn/kofi3.png?v=0' border='0' alt='Kauf mir einen Kaffee auf ko-fi.com' />
+
+<div id="top"></div>
+
+> Durch die Nutzung dieses Repositories oder jeglichen damit verbundenen Code stimmen Sie dem [Rechtshinweis](LEGAL_NOTICE.md) zu. Der Autor ist nicht verantwortlich für Kopien, Forks, erneute Uploads durch andere Benutzer oder sonstige mit GPT4Free verbundene Aktivitäten. Dies ist das einzige Konto und Repository des Autors. Um Identitätsdiebstahl oder unverantwortliche Handlungen zu verhindern, halten Sie sich bitte an die GNU GPL-Lizenz, die dieses Repository verwendet.
+
+```sh
+pip install -U g4f
+```
+
+
+## 🆕 Was gibt es Neues
+
+- Tritt unserem Telegram-Kanal bei: [t.me/g4f_channel](https://telegram.me/g4f_channel)
+- Tritt unserer Discord-Gruppe bei: [discord.gg/XfybzPXPH5](https://discord.gg/XfybzPXPH5)
+- Erkunde die g4f-Dokumentation (unvollständig): [g4f.mintlify.app](https://g4f.mintlify.app) | Trage zur Dokumentation bei: [github.com/xtekky/gpt4free-docs](https://github.com/xtekky/gpt4free-docs)
+
+
+## 📚 Inhaltsverzeichnis
+
+- [🆕 Was ist neu](#-was-ist-neu)
+- [📚 Inhaltsverzeichnis](#-inhaltsverzeichnis)
+- [🛠️ Erste Schritte](#️-erste-schritte)
+    - [Voraussetzungen:](#voraussetzungen)
+    - [Projekt einrichten:](#projekt-einrichten)
+      - [Installation über PyPi](#installation-über-pypi)
+      - [oder](#oder)
+      - [Einrichten mit Docker:](#einrichten-mit-docker)
+- [💡 Verwendung](#-verwendung)
+  - [Das `g4f` Paket](#das-g4f-paket)
+    - [ChatCompletion](#chatcompletion)
+      - [Vervollständigung](#vervollständigung)
+      - [Anbieter](#anbieter)
+      - [Cookies erforderlich](#cookies-erforderlich)
+      - [Async-Unterstützung](#async-unterstützung)
+      - [Proxy- und Timeout-Unterstützung](#proxy-und-timeout-unterstützung)
+  - [Interference openai-proxy API (Verwendung mit openai Python-Paket)](#interference-openai-proxy-api-verwendung-mit-openai-python-paket)
+    - [API von PyPi-Paket ausführen](#api-von-pypi-paket-ausführen)
+    - [API von Repository ausführen](#api-von-repository-ausführen)
+- [🚀 Anbieter und Modelle](#-anbieter-und-modelle)
+  - [GPT-4](#gpt-4)
+  - [GPT-3.5](#gpt-35)
+  - [Andere](#andere)
+  - [Modelle](#modelle)
+- [🔗 Verwandte GPT4Free-Projekte](#-verwandte-gpt4free-projekte)
+- [🤝 Mitwirken](#-mitwirken)
+    - [Anbieter mit KI-Tool erstellen](#anbieter-mit-ki-tool-erstellen)
+    - [Anbieter erstellen](#anbieter-erstellen)
+- [🙌 Mitwirkende](#-mitwirkende)
+- [©️ Urheberrecht](#️-urheberrecht)
+- [⭐ Sternenhistorie](#-sternenhistorie)
+- [📄 Lizenz](#-lizenz)
+
+
+## 🛠️ Erste Schritte
+
+#### Voraussetzungen:
+
+1. [Python herunterladen und installieren](https://www.python.org/downloads/) (Version 3.10+ wird empfohlen).
+
+#### Projekt einrichten:
+
+##### Installation über pypi
+
+```
+pip install -U g4f
+```
+
+##### oder
+
+1. Klonen Sie das GitHub-Repository:
+
+```
+git clone https://github.com/xtekky/gpt4free.git
+```
+
+2. Navigieren Sie zum Projektverzeichnis:
+
+```
+cd gpt4free
+```
+
+3. (Empfohlen) Erstellen Sie eine Python-Virtual-Umgebung:
+Sie können der [Python-Offiziellen Dokumentation](https://docs.python.org/3/tutorial/venv.html) für virtuelle Umgebungen folgen.
+
+```
+python3 -m venv venv
+```
+
+4. Aktivieren Sie die virtuelle Umgebung:
+   - Unter Windows:
+   ```
+   .\venv\Scripts\activate
+   ```
+   - Unter macOS und Linux:
+   ```
+   source venv/bin/activate
+   ```
+5. Installieren Sie die erforderlichen Python-Pakete aus `requirements.txt`:
+
+```
+pip install -r requirements.txt
+```
+
+6. Erstellen Sie eine Datei `test.py` im Stammverzeichnis und beginnen Sie mit der Verwendung des Repositories. Weitere Anweisungen finden Sie unten
+
+```py
+import g4f
+
+...
+```
+
+##### Einrichten mit Docker:
+
+Wenn Docker installiert ist, können Sie das Projekt ohne manuelle Installation von Abhängigkeiten einfach einrichten und ausführen.
+
+1. Stellen Sie zunächst sicher, dass sowohl Docker als auch Docker Compose installiert sind.
+
+   - [Docker installieren](https://docs.docker.com/get-docker/)
+   - [Docker Compose installieren](https://docs.docker.com/compose/install/)
+
+2. Klonen Sie das GitHub-Repo:
+
+```bash
+git clone https://github.com/xtekky/gpt4free.git
+```
+
+3. Navigieren Sie zum Projektverzeichnis:
+
+```bash
+cd gpt4free
+```
+
+4. Erstellen Sie das Docker-Image:
+
+```bash
+docker-compose build
+```
+
+5. Starten Sie den Dienst mit Docker Compose:
+
+```bash
+docker-compose up
+```
+
+Ihr Server wird jetzt unter `http://localhost:1337` ausgeführt. Sie können mit der API interagieren oder Ihre Tests wie gewohnt ausführen.
+
+Um die Docker-Container zu stoppen, führen Sie einfach aus:
+
+```bash
+docker-compose down
+```
+
+> [!Note]
+> Wenn Sie Docker verwenden, werden alle Änderungen, die Sie an Ihren lokalen Dateien vornehmen, im Docker-Container durch die Volumenabbildung in der `docker-compose.yml`-Datei widergespiegelt. Wenn Sie jedoch Abhängigkeiten hinzufügen oder entfernen, müssen Sie das Docker-Image mit `docker-compose build` neu erstellen.
+
+
+## 💡 Verwendung
+
+### Das `g4f` Paket
+
+#### ChatCompletion
+
+```python
+import g4f
+
+g4f.debug.logging = True  # Aktiviere das Protokollieren
+g4f.check_version = False  # Deaktiviere die automatische Versionsüberprüfung
+print(g4f.version)  # Überprüfe die Version
+print(g4f.Provider.Ails.params)  # Unterstützte Argumente
+
+# Automatische Auswahl des Anbieters
+
+# Gestreamte Vervollständigung
+response = g4f.ChatCompletion.create(
+    model="gpt-3.5-turbo",
+    messages=[{"role": "user", "content": "Hallo"}],
+    stream=True,
+)
+
+for message in response:
+    print(message, flush=True, end='')
+
+# Normale Antwort
+response = g4f.ChatCompletion.create(
+    model=g4f.models.gpt_4,
+    messages=[{"role": "user", "content": "Hallo"}],
+)  # Alternative Modellkonfiguration
+
+print(response)
+```
+
+##### Completion
+
+```python
+import g4f
+
+erlaubte_modelle = [
+    'code-davinci-002',
+    'text-ada-001',
+    'text-babbage-001',
+    'text-curie-001',
+    'text-davinci-002',
+    'text-davinci-003'
+]
+
+response = g4f.Completion.create(
+    model='text-davinci-003',
+    prompt='sage, dass dies ein Test ist'
+)
+
+print(response)
+```
+
+##### Anbieter
+
+```python
+import g4f
+
+from g4f.Provider import (
+    AItianhu,
+    Aichat,
+    Bard,
+    Bing,
+    ChatBase,
+    ChatgptAi,
+    OpenaiChat,
+    Vercel,
+    You,
+    Yqcloud,
+)
+
+# Festlegen des Anbieters
+response = g4f.ChatCompletion.create(
+    model="gpt-3.5-turbo",
+    provider=g4f.Provider.Aichat,
+    messages=[{"role": "user", "content": "Hallo"}],
+    stream=True,
+)
+
+for message in response:
+    print(message)
+```
+
+##### Verwendung des Browsers
+
+Einige Anbieter verwenden einen Browser, um den Bot-Schutz zu umgehen.
+Sie verwenden den Selenium-Webtreiber, um den Browser zu steuern.
+Die Browsereinstellungen und die Anmeldedaten werden in einem benutzerdefinierten Verzeichnis gespeichert.
+Wenn der Headless-Modus aktiviert ist, werden die Browserfenster unsichtbar geladen.
+Aus Leistungsgründen wird empfohlen, die Browserinstanzen wiederzuverwenden
+und sie am Ende selbst zu schließen:
+
+```python
+import g4f
+from undetected_chromedriver import Chrome, ChromeOptions
+from g4f.Provider import (
+    Bard,
+    Poe,
+    AItianhuSpace,
+    MyShell,
+    Phind,
+    PerplexityAi,
+)
+
+options = ChromeOptions()
+options.add_argument("--incognito")
+browser = Chrome(options=options, headless=True)
+for idx in range(10):
+    response = g4f.ChatCompletion.create(
+        model=g4f.models.default,
+        provider=g4f.Provider.Phind,
+        messages=[{"role": "user", "content": "Schlage mir einen Namen vor."}],
+        browser=browser
+    )
+    print(f"{idx}:", response)
+browser.quit()
+```
+
+##### Erforderliche Cookies
+
+Cookies sind für die ordnungsgemäße Funktion einiger Dienstanbieter unerlässlich. Es ist unerlässlich, eine aktive Sitzung aufrechtzuerhalten, die in der Regel durch das Anmelden in Ihrem Konto erreicht wird.
+
+Wenn Sie das g4f-Paket lokal ausführen, ruft das Paket automatisch Cookies aus Ihrem Webbrowser ab, indem es die `get_cookies`-Funktion verwendet. Wenn Sie es jedoch nicht lokal ausführen, müssen Sie die Cookies manuell bereitstellen, indem Sie sie als Parameter unter Verwendung des `cookies`-Parameters übergeben.
+
+```python
+import g4f
+
+from g4f.Provider import (
+    Bing,
+    HuggingChat,
+    OpenAssistant,
+)
+
+# Verwendung
+response = g4f.ChatCompletion.create(
+    model=g4f.models.default,
+    messages=[{"role": "user", "content": "Hallo"}],
+    provider=Bing,
+    #cookies=g4f.get_cookies(".google.com"),
+    cookies={"cookie_name": "value", "cookie_name2": "value2"},
+    auth=True
+)
+```
+
+##### Unterstützung für asynchrone Ausführung
+
+Um die Geschwindigkeit und Gesamtleistung zu verbessern, führen Sie Anbieter asynchron aus. Die Gesamtausführungszeit wird durch die Dauer der langsamsten Anbieterausführung bestimmt.
+
+```python
+import g4f
+import asyncio
+
+_providers = [
+    g4f.Provider.Aichat,
+    g4f.Provider.ChatBase,
+    g4f.Provider.Bing,
+    g4f.Provider.GptGo,
+    g4f.Provider.You,
+    g4f.Provider.Yqcloud,
+]
+
+async def run_provider(provider: g4f.Provider.BaseProvider):
+    try:
+        response = await g4f.ChatCompletion.create_async(
+            model=g4f.models.default,
+            messages=[{"role": "user", "content": "Hallo"}],
+            provider=provider,
+        )
+        print(f"{provider.__name__}:", response)
+    except Exception as e:
+        print(f"{provider.__name__}:", e)
+        
+async def run_all():
+    calls = [
+        run_provider(provider) for provider in _providers
+    ]
+    await asyncio.gather(*calls)
+
+asyncio.run(run_all())
+```
+
+##### Unterstützung für Proxy und Timeout
+
+Alle Anbieter unterstützen das Angeben eines Proxy und das Erhöhen des Timeouts in den Erstellungsfunktionen.
+
+```python
+import g4f
+
+response = g4f.ChatCompletion.create(
+    model=g4f.models.default,
+    messages=[{"role": "user", "content": "Hallo"}],
+    proxy="http://host:port",
+    # oder socks5://user:pass@host:port
+    timeout=120,  # in Sekunden
+)
+
+print(f"Ergebnis:", response)
+```
+
+### Interference openai-proxy API (Verwendung mit dem openai Python-Paket)
+
+#### Führen Sie die Interference API aus dem PyPi-Paket aus
+
+```python
+from g4f.api import run_api
+
+run_api()
+```
+
+#### Führen Sie die Interference API aus dem Repository aus
+
+Wenn Sie die Einbettungsfunktion verwenden möchten, benötigen Sie einen Hugging Face-Token. Sie können einen unter [Hugging Face Tokens](https://huggingface.co/settings/tokens) erhalten. Stellen Sie sicher, dass Ihre Rolle auf Schreiben eingestellt ist. Wenn Sie Ihren Token haben, verwenden Sie ihn einfach anstelle des OpenAI-API-Schlüssels.
+
+Server ausführen:
+
+```sh
+g4f api
+```
+
+oder
+
+```sh
+python -m g4f.api
+```
+
+```python
+import openai
+
+# Setzen Sie Ihren Hugging Face-Token als API-Schlüssel, wenn Sie Einbettungen verwenden
+# Wenn Sie keine Einbettungen verwenden, lassen Sie es leer
+openai.api_key = "IHR_HUGGING_FACE_TOKEN"  # Ersetzen Sie dies durch Ihren tatsächlichen Token
+
+# Setzen Sie die API-Basis-URL, falls erforderlich, z.B. für eine lokale Entwicklungsumgebung
+openai.api_base = "http://localhost:1337/v1"
+
+def main():
+    chat_completion = openai.ChatCompletion.create(
+        model="gpt-3.5-turbo",
+        messages=[{"role": "user", "content": "schreibe ein Gedicht über einen Baum"}],
+        stream=True,
+    )
+
+    if isinstance(chat_completion, dict):
+        # Nicht gestreamt
+        print(chat_completion.choices[0].message.content)
+    else:
+        # Gestreamt
+        for token in chat_completion:
+            content = token["choices"][0]["delta"].get("content")
+            if content is not None:
+                print(content, end="", flush=True)
+
+if __name__ == "__main__":
+    main()
+```
+
+## 🚀 Anbieter und Modelle
+
+### GPT-4
+
+| Website | Provider | GPT-3.5 | GPT-4 | Stream | Status | Auth |
+| ------  | -------  | ------- | ----- | ------ | ------ | ---- |
+| [bing.com](https://bing.com/chat) | `g4f.Provider.Bing` | ❌ | ✔️ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [chat.geekgpt.org](https://chat.geekgpt.org) | `g4f.Provider.GeekGpt` | ✔️ | ✔️ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [gptchatly.com](https://gptchatly.com) | `g4f.Provider.GptChatly` | ✔️ | ✔️ | ❌ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [liaobots.site](https://liaobots.site) | `g4f.Provider.Liaobots` | ✔️ | ✔️ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [www.phind.com](https://www.phind.com) | `g4f.Provider.Phind` | ❌ | ✔️ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [raycast.com](https://raycast.com) | `g4f.Provider.Raycast` | ✔️ | ✔️ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ✔️ |
+
+### GPT-3.5
+
+| Website | Provider | GPT-3.5 | GPT-4 | Stream | Status | Auth |
+| ------  | -------  | ------- | ----- | ------ | ------ | ---- |
+| [www.aitianhu.com](https://www.aitianhu.com) | `g4f.Provider.AItianhu` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [chat3.aiyunos.top](https://chat3.aiyunos.top/) | `g4f.Provider.AItianhuSpace` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [e.aiask.me](https://e.aiask.me) | `g4f.Provider.AiAsk` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [chat-gpt.org](https://chat-gpt.org/chat) | `g4f.Provider.Aichat` | ✔️ | ❌ | ❌ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [www.chatbase.co](https://www.chatbase.co) | `g4f.Provider.ChatBase` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [chatforai.store](https://chatforai.store) | `g4f.Provider.ChatForAi` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [chatgpt.ai](https://chatgpt.ai) | `g4f.Provider.ChatgptAi` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [chatgptx.de](https://chatgptx.de) | `g4f.Provider.ChatgptX` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [chat-shared2.zhile.io](https://chat-shared2.zhile.io) | `g4f.Provider.FakeGpt` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [freegpts1.aifree.site](https://freegpts1.aifree.site/) | `g4f.Provider.FreeGpt` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [gptalk.net](https://gptalk.net) | `g4f.Provider.GPTalk` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [ai18.gptforlove.com](https://ai18.gptforlove.com) | `g4f.Provider.GptForLove` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [gptgo.ai](https://gptgo.ai) | `g4f.Provider.GptGo` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [hashnode.com](https://hashnode.com) | `g4f.Provider.Hashnode` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [app.myshell.ai](https://app.myshell.ai/chat) | `g4f.Provider.MyShell` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [noowai.com](https://noowai.com) | `g4f.Provider.NoowAi` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [chat.openai.com](https://chat.openai.com) | `g4f.Provider.OpenaiChat` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ✔️ |
+| [theb.ai](https://theb.ai) | `g4f.Provider.Theb` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ✔️ |
+| [sdk.vercel.ai](https://sdk.vercel.ai) | `g4f.Provider.Vercel` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [you.com](https://you.com) | `g4f.Provider.You` | ✔️ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [chat9.yqcloud.top](https://chat9.yqcloud.top/) | `g4f.Provider.Yqcloud` | ✔️ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [chat.acytoo.com](https://chat.acytoo.com) | `g4f.Provider.Acytoo` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [aibn.cc](https://aibn.cc) | `g4f.Provider.Aibn` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [ai.ls](https://ai.ls) | `g4f.Provider.Ails` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [chatgpt4online.org](https://chatgpt4online.org) | `g4f.Provider.Chatgpt4Online` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [chat.chatgptdemo.net](https://chat.chatgptdemo.net) | `g4f.Provider.ChatgptDemo` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [chatgptduo.com](https://chatgptduo.com) | `g4f.Provider.ChatgptDuo` | ✔️ | ❌ | ❌ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [chatgptfree.ai](https://chatgptfree.ai) | `g4f.Provider.ChatgptFree` | ✔️ | ❌ | ❌ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [chatgptlogin.ai](https://chatgptlogin.ai) | `g4f.Provider.ChatgptLogin` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [cromicle.top](https://cromicle.top) | `g4f.Provider.Cromicle` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [gptgod.site](https://gptgod.site) | `g4f.Provider.GptGod` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [opchatgpts.net](https://opchatgpts.net) | `g4f.Provider.Opchatgpts` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+| [chat.ylokh.xyz](https://chat.ylokh.xyz) | `g4f.Provider.Ylokh` | ✔️ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ❌ |
+
+### Andere
+
+| Website | Provider | GPT-3.5 | GPT-4 | Stream | Status | Auth |
+| ------  | -------  | ------- | ----- | ------ | ------ | ---- |
+| [bard.google.com](https://bard.google.com) | `g4f.Provider.Bard` | ❌ | ❌ | ❌ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ✔️ |
+| [deepinfra.com](https://deepinfra.com) | `g4f.Provider.DeepInfra` | ❌ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ❌ |
+| [huggingface.co](https://huggingface.co/chat) | `g4f.Provider.HuggingChat` | ❌ | ❌ | ✔️ | ![Active](https://img.shields.io/badge/Active-brightgreen) | ✔️ |
+| [www.llama2.ai](https://www.llama2.ai) | `g4f.Provider.Llama2` | ❌ | ❌ | ✔️ | ![Unknown](https://img.shields.io/badge/Unknown-grey) | ❌ |
+| [open-assistant.io](https://open-assistant.io/chat) | `g4f.Provider.OpenAssistant` | ❌ | ❌ | ✔️ | ![Inactive](https://img.shields.io/badge/Inactive-red) | ✔️ |
+
+### Modelle
+
+| Model                                   | Base Provider | Provider            | Website                                     |
+| --------------------------------------- | ------------- | ------------------- | ------------------------------------------- |
+| palm                                    | Google        | g4f.Provider.Bard   | [bard.google.com](https://bard.google.com/) |
+| h2ogpt-gm-oasst1-en-2048-falcon-7b-v3   | Hugging Face  | g4f.Provider.H2o    | [www.h2o.ai](https://www.h2o.ai/)           |
+| h2ogpt-gm-oasst1-en-2048-falcon-40b-v1  | Hugging Face  | g4f.Provider.H2o    | [www.h2o.ai](https://www.h2o.ai/)           |
+| h2ogpt-gm-oasst1-en-2048-open-llama-13b | Hugging Face  | g4f.Provider.H2o    | [www.h2o.ai](https://www.h2o.ai/)           |
+| claude-instant-v1                       | Anthropic     | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| claude-v1                               | Anthropic     | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| claude-v2                               | Anthropic     | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| command-light-nightly                   | Cohere        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| command-nightly                         | Cohere        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| gpt-neox-20b                            | Hugging Face  | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| oasst-sft-1-pythia-12b                  | Hugging Face  | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| oasst-sft-4-pythia-12b-epoch-3.5        | Hugging Face  | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| santacoder                              | Hugging Face  | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| bloom                                   | Hugging Face  | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| flan-t5-xxl                             | Hugging Face  | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| code-davinci-002                        | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| gpt-3.5-turbo-16k                       | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| gpt-3.5-turbo-16k-0613                  | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| gpt-4-0613                              | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| text-ada-001                            | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| text-babbage-001                        | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| text-curie-001                          | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| text-davinci-002                        | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| text-davinci-003                        | OpenAI        | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| llama13b-v2-chat                        | Replicate     | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+| llama7b-v2-chat                         | Replicate     | g4f.Provider.Vercel | [sdk.vercel.ai](https://sdk.vercel.ai/)     |
+
+
+## 🔗 Verwandte GPT4Free-Projekte
+
+<table>
+  <thead align="center">
+    <tr border: none;>
+      <td><b>🎁 Projects</b></td>
+      <td><b>⭐ Stars</b></td>
+      <td><b>📚 Forks</b></td>
+      <td><b>🛎 Issues</b></td>
+      <td><b>📬 Pull requests</b></td>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td><a href="https://github.com/xtekky/gpt4free"><b>gpt4free</b></a></td>
+      <td><a href="https://github.com/xtekky/gpt4free/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/xtekky/gpt4free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xtekky/gpt4free/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/xtekky/gpt4free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xtekky/gpt4free/issues"><img alt="Issues" src="https://img.shields.io/github/issues/xtekky/gpt4free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xtekky/gpt4free/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/xtekky/gpt4free?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+      <td><a href="https://github.com/xiangsx/gpt4free-ts"><b>gpt4free-ts</b></a></td>
+      <td><a href="https://github.com/xiangsx/gpt4free-ts/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/xiangsx/gpt4free-ts?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xiangsx/gpt4free-ts/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/xiangsx/gpt4free-ts?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xiangsx/gpt4free-ts/issues"><img alt="Issues" src="https://img.shields.io/github/issues/xiangsx/gpt4free-ts?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xiangsx/gpt4free-ts/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/xiangsx/gpt4free-ts?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+     <tr>
+      <td><a href="https://github.com/zukixa/cool-ai-stuff/"><b>Free AI API's & Potential Providers List</b></a></td>
+      <td><a href="https://github.com/zukixa/cool-ai-stuff/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/zukixa/cool-ai-stuff?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/zukixa/cool-ai-stuff/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/zukixa/cool-ai-stuff?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/zukixa/cool-ai-stuff/issues"><img alt="Issues" src="https://img.shields.io/github/issues/zukixa/cool-ai-stuff?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/zukixa/cool-ai-stuff/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/zukixa/cool-ai-stuff?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+    <tr>
+    <tr>
+      <td><a href="https://github.com/xtekky/chatgpt-clone"><b>ChatGPT-Clone</b></a></td>
+      <td><a href="https://github.com/xtekky/chatgpt-clone/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/xtekky/chatgpt-clone?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xtekky/chatgpt-clone/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/xtekky/chatgpt-clone?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xtekky/chatgpt-clone/issues"><img alt="Issues" src="https://img.shields.io/github/issues/xtekky/chatgpt-clone?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/xtekky/chatgpt-clone/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/xtekky/chatgpt-clone?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+    <tr>
+      <td><a href="https://github.com/mishalhossin/Discord-Chatbot-Gpt4Free"><b>ChatGpt Discord Bot</b></a></td>
+      <td><a href="https://github.com/mishalhossin/Discord-Chatbot-Gpt4Free/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/mishalhossin/Discord-Chatbot-Gpt4Free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/mishalhossin/Discord-Chatbot-Gpt4Free/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/mishalhossin/Discord-Chatbot-Gpt4Free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/mishalhossin/Discord-Chatbot-Gpt4Free/issues"><img alt="Issues" src="https://img.shields.io/github/issues/mishalhossin/Discord-Chatbot-Gpt4Free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/mishalhossin/Coding-Chatbot-Gpt4Free/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/mishalhossin/Discord-Chatbot-Gpt4Free?style=flat-square&labelColor=343b41"/></a></td>
+<tr>
+  <td><a href="https://github.com/SamirXR/Nyx-Bot"><b>Nyx-Bot (Discord)</b></a></td>
+  <td><a href="https://github.com/SamirXR/Nyx-Bot/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/SamirXR/Nyx-Bot?style=flat-square&labelColor=343b41"/></a></td>
+  <td><a href="https://github.com/SamirXR/Nyx-Bot/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/SamirXR/Nyx-Bot?style=flat-square&labelColor=343b41"/></a></td>
+  <td><a href="https://github.com/SamirXR/Nyx-Bot/issues"><img alt="Issues" src="https://img.shields.io/github/issues/SamirXR/Nyx-Bot?style=flat-square&labelColor=343b41"/></a></td>
+  <td><a href="https://github.com/SamirXR/Nyx-Bot/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/SamirXR/Nyx-Bot?style=flat-square&labelColor=343b41"/></a></td>
+</tr>
+    </tr>
+    <tr>
+      <td><a href="https://github.com/MIDORIBIN/langchain-gpt4free"><b>LangChain gpt4free</b></a></td>
+      <td><a href="https://github.com/MIDORIBIN/langchain-gpt4free/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/MIDORIBIN/langchain-gpt4free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/MIDORIBIN/langchain-gpt4free/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/MIDORIBIN/langchain-gpt4free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/MIDORIBIN/langchain-gpt4free/issues"><img alt="Issues" src="https://img.shields.io/github/issues/MIDORIBIN/langchain-gpt4free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/MIDORIBIN/langchain-gpt4free/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/MIDORIBIN/langchain-gpt4free?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+    <tr>
+      <td><a href="https://github.com/HexyeDEV/Telegram-Chatbot-Gpt4Free"><b>ChatGpt Telegram Bot</b></a></td>
+      <td><a href="https://github.com/HexyeDEV/Telegram-Chatbot-Gpt4Free/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/HexyeDEV/Telegram-Chatbot-Gpt4Free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/HexyeDEV/Telegram-Chatbot-Gpt4Free/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/HexyeDEV/Telegram-Chatbot-Gpt4Free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/HexyeDEV/Telegram-Chatbot-Gpt4Free/issues"><img alt="Issues" src="https://img.shields.io/github/issues/HexyeDEV/Telegram-Chatbot-Gpt4Free?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/HexyeDEV/Telegram-Chatbot-Gpt4Free/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/HexyeDEV/Telegram-Chatbot-Gpt4Free?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+        <tr>
+      <td><a href="https://github.com/Lin-jun-xiang/chatgpt-line-bot"><b>ChatGpt Line Bot</b></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/chatgpt-line-bot/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/Lin-jun-xiang/chatgpt-line-bot?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/chatgpt-line-bot/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/Lin-jun-xiang/chatgpt-line-bot?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/chatgpt-line-bot/issues"><img alt="Issues" src="https://img.shields.io/github/issues/Lin-jun-xiang/chatgpt-line-bot?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/chatgpt-line-bot/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/Lin-jun-xiang/chatgpt-line-bot?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+    <tr>
+      <td><a href="https://github.com/Lin-jun-xiang/action-translate-readme"><b>Action Translate Readme</b></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/action-translate-readme/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/Lin-jun-xiang/action-translate-readme?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/action-translate-readme/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/Lin-jun-xiang/action-translate-readme?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/action-translate-readme/issues"><img alt="Issues" src="https://img.shields.io/github/issues/Lin-jun-xiang/action-translate-readme?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/action-translate-readme/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/Lin-jun-xiang/action-translate-readme?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+    <tr>
+      <td><a href="https://github.com/Lin-jun-xiang/docGPT-streamlit"><b>Langchain Document GPT</b></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/docGPT-streamlit/stargazers"><img alt="Stars" src="https://img.shields.io/github/stars/Lin-jun-xiang/docGPT-streamlit?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/docGPT-streamlit/network/members"><img alt="Forks" src="https://img.shields.io/github/forks/Lin-jun-xiang/docGPT-streamlit?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/docGPT-streamlit/issues"><img alt="Issues" src="https://img.shields.io/github/issues/Lin-jun-xiang/docGPT-streamlit?style=flat-square&labelColor=343b41"/></a></td>
+      <td><a href="https://github.com/Lin-jun-xiang/docGPT-streamlit/pulls"><img alt="Pull Requests" src="https://img.shields.io/github/issues-pr/Lin-jun-xiang/docGPT-streamlit?style=flat-square&labelColor=343b41"/></a></td>
+    </tr>
+  </tbody>
+</table>
+
+
+
+## 🤝 Mitwirken
+
+#### Erstellen Sie einen Anbieter mit AI-Tool
+
+Rufen Sie im Terminal das Skript `create_provider.py` auf:
+```bash
+python etc/tool/create_provider.py
+```
+1. Geben Sie Ihren Namen für den neuen Anbieter ein.
+2. Kopieren Sie den `cURL`-Befehl aus den Entwicklertools Ihres Browsers und fügen Sie ihn ein.
+3. Lassen Sie die KI den Anbieter für Sie erstellen.
+4. Passen Sie den Anbieter nach Ihren Bedürfnissen an.
+
+#### Anbieter erstellen
+
+1. Überprüfen Sie die aktuelle [Liste potenzieller Anbieter](https://github.com/zukixa/cool-ai-stuff#ai-chat-websites) oder finden Sie Ihre eigene Anbieterquelle!
+2. Erstellen Sie eine neue Datei in [g4f/Provider](./g4f/Provider) mit dem Namen des Anbieters.
+3. Implementieren Sie eine Klasse, die von [BaseProvider](./g4f/Provider/base_provider.py) erbt.
+
+```py
+from __future__ import annotations
+
+from ..typing import AsyncResult, Messages
+from .base_provider import AsyncGeneratorProvider
+
+class HogeService(AsyncGeneratorProvider):
+    url                   = "https://chat-gpt.com"
+    supports_gpt_35_turbo = True
+    working               = True
+
+    @classmethod
+    async def create_async_generator(
+        cls,
+        model: str,
+        messages: Messages,
+        proxy: str = None,
+        **kwargs
+    ) -> AsyncResult:
+        yield ""
+```
+
+4. Hier können Sie die Einstellungen anpassen, zum Beispiel, wenn die Website Streaming unterstützt, setzen Sie `supports_stream` auf `True`...
+5. Schreiben Sie Code, um den Anbieter in `create_async_generator` anzufordern und die Antwort mit `yield` zurückzugeben, selbst wenn es sich um eine einmalige Antwort handelt. Zögern Sie nicht, sich bei anderen Anbietern inspirieren zu lassen.
+6. Fügen Sie den Namen des Anbieters in [`g4f/Provider/__init__.py`](./g4f/Provider/__init__.py) hinzu.
+
+```py
+from .HogeService import HogeService
+
+__all__ = [
+  HogeService,
+]
+```
+
+7. Sie sind fertig! Testen Sie den Anbieter, indem Sie ihn aufrufen:
+
+```py
+import g4f
+
+response = g4f.ChatCompletion.create(model='gpt-3.5-turbo', provider=g4f.Provider.PROVIDERNAME,
+                                    messages=[{"role": "user", "content": "test"}], stream=g4f.Provider.PROVIDERNAME.supports_stream)
+
+for message in response:
+    print(message, flush=True, end='')
+```
+
+
+## 🙌 Mitwirkende
+
+Eine Liste der Mitwirkenden ist [hier](https://github.com/xtekky/gpt4free/graphs/contributors) verfügbar.
+Die Datei [`Vercel.py`](https://github.com/xtekky/gpt4free/blob/main/g4f/Provider/Vercel.py) enthält Code von [vercel-llm-api](https://github.com/ading2210/vercel-llm-api) von [@ading2210](https://github.com/ading2210), der unter der [GNU GPL v3](https://www.gnu.org/licenses/gpl-3.0.txt) lizenziert ist.
+Top 1 Mitwirkender: [@hlohaus](https://github.com/hlohaus)
+
+## ©️ Urheberrecht
+
+This program is licensed under the [GNU GPL v3](https://www.gnu.org/licenses/gpl-3.0.txt)
+
+```
+xtekky/gpt4free: Copyright (C) 2023 xtekky
+
+This program is free software: you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation, either version 3 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program.  If not, see <https://www.gnu.org/licenses/>.
+```
+
+## ⭐ Sternenverlauf
+
+<a href="https://github.com/xtekky/gpt4free/stargazers">
+        <img width="500" alt="Star History Chart" src="https://api.star-history.com/svg?repos=xtekky/gpt4free&type=Date">
+</a>
+
+
+## 📄 Lizenz
+
+<table>
+  <tr>
+    <td>
+      <p align="center"> <img src="https://upload.wikimedia.org/wikipedia/commons/thumb/9/93/GPLv3_Logo.svg/1200px-GPLv3_Logo.svg.png" width="80%"></img>
+    </td>
+    <td>
+      <img src="https://img.shields.io/badge/Lizenz-GNU_GPL_v3.0-rot.svg"/> <br>
+      Dieses Projekt steht unter der <a href="./LICENSE">GNU_GPL_v3.0-Lizenz</a>.
+    </td>
+  </tr>
+</table>
+
+<p align="right">(<a href="#top">🔼 Zurück nach oben</a>)</p>

README.md

@@ -1,3 +1,7 @@
+<a href="./README-DE.md">
+    <img src="https://img.shields.io/badge/öffnen in-🇩🇪 deutsch-bleu.svg" alt="Öffnen en DE">
+</a>
+
 ![248433934-7886223b-c1d1-4260-82aa-da5741f303bb](https://github.com/xtekky/gpt4free/assets/98614666/ea012c87-76e0-496a-8ac4-e2de090cc6c9)
 
 <a href='https://ko-fi.com/xtekky' target='_blank'><img height='35' style='border:0px;height:46px;' src='https://az743702.vo.msecnd.net/cdn/kofi3.png?v=0' border='0' alt='Buy Me a Coffee at ko-fi.com' />
@@ -245,12 +249,7 @@ for message in response:
 
 ##### Using Browser
 
-Some providers using a browser to bypass the bot protection.
-They using the selenium webdriver to control the browser.
-The browser settings and the login data are saved in a custom directory.
-If the headless mode is enabled, the browser windows are loaded invisibly.
-For performance reasons, it is recommended to reuse the browser instances
-and close them yourself at the end:
+Some providers using a a browser to bypass the bot protection. They using the selenium webdriver to control the browser. The browser settings and the login data are saved in a custom directory. If the headless mode is enabled, the browser windows are loaded invisibly. For performance reasons, it is recommended to reuse the browser instances and close them yourself at the end:
 
 ```python
 import g4f
@@ -266,16 +265,16 @@ from g4f.Provider import (
 
 options = ChromeOptions()
 options.add_argument("--incognito");
-browser = Chrome(options=options, headless=True)
+webdriver = Chrome(options=options, headless=True)
 for idx in range(10):
     response = g4f.ChatCompletion.create(
         model=g4f.models.default,
         provider=g4f.Provider.Phind,
         messages=[{"role": "user", "content": "Suggest me a name."}],
-        browser=browser
+        webdriver=webdriver
     )
     print(f"{idx}:", response)
-browser.quit()
+webdriver.quit()
 ```
 
 ##### Cookies Required
@@ -605,7 +604,7 @@ if __name__ == "__main__":
 
 #### Create Provider with AI Tool
 
-Call in your terminal the "create_provider" script:
+Call in your terminal the `create_provider.py` script:
 ```bash
 python etc/tool/create_provider.py
 ```
@@ -628,8 +627,8 @@ from .base_provider import AsyncGeneratorProvider
 
 class HogeService(AsyncGeneratorProvider):
     url                   = "https://chat-gpt.com"
-    supports_gpt_35_turbo = True
     working               = True
+    supports_gpt_35_turbo = True
 
     @classmethod
     async def create_async_generator(
@@ -644,7 +643,7 @@ class HogeService(AsyncGeneratorProvider):
 
 4. Here, you can adjust the settings, for example, if the website does support streaming, set `supports_stream` to `True`...
 5. Write code to request the provider in `create_async_generator` and `yield` the response, _even if_ it's a one-time response, do not hesitate to look at other providers for inspiration
-6. Add the Provider Name in [g4f/Provider/**init**.py](./g4f/Provider/__init__.py)
+6. Add the Provider Name in [`g4f/Provider/__init__.py`](./g4f/Provider/__init__.py)
 
 ```py
 from .HogeService import HogeService
@@ -708,7 +707,7 @@ along with this program.  If not, see <https://www.gnu.org/licenses/>.
     </td>
     <td> 
       <img src="https://img.shields.io/badge/License-GNU_GPL_v3.0-red.svg"/> <br> 
-This project is licensed under <a href="./LICENSE">GNU_GPL_v3.0</a>. <img width=2300/>
+This project is licensed under <a href="./LICENSE">GNU_GPL_v3.0</a>.
     </td>
   </tr>
 </table>

etc/tool/translate_readme.py

@@ -0,0 +1,88 @@
+
+import sys
+from pathlib import Path
+import asyncio
+
+sys.path.append(str(Path(__file__).parent.parent.parent))
+
+import g4f
+g4f.debug.logging = True
+from g4f.debug import access_token
+provider = g4f.Provider.OpenaiChat
+
+iso = "GE"
+language = "german"
+translate_prompt = f"""
+Translate this markdown document to {language}.
+Don't translate or change inline code examples.
+```md
+"""
+keep_note = "Keep this: [!Note] as [!Note].\n"
+blacklist = [
+    '## ©️ Copyright',
+    '## 🚀 Providers and Models',
+    '## 🔗 Related GPT4Free Projects'
+]
+whitelist = [
+    "### Other",
+    "### Models"
+]
+
+def read_text(text):
+    start = end = 0
+    new = text.strip().split('\n')
+    for i, line in enumerate(new):
+        if line.startswith('```'):
+            if not start:
+                start = i + 1
+            end = i
+    return '\n'.join(new[start:end]).strip()
+
+async def translate(text):
+    prompt = translate_prompt + text.strip() + '\n```'
+    if "[!Note]" in text:
+        prompt = keep_note + prompt
+    result = read_text(await provider.create_async(
+        model="",
+        messages=[{"role": "user", "content": prompt}],
+        access_token=access_token
+    ))
+    if text.endswith("```") and not result.endswith("```"):
+        result += "\n```"
+    return result
+
+async def translate_part(part, i):
+    blacklisted = False
+    for headline in blacklist:
+        if headline in part:
+            blacklisted = True
+    if blacklisted:
+        lines = part.split('\n')
+        lines[0] = await translate(lines[0])
+        part = '\n'.join(lines)
+        for trans in whitelist:
+            if trans in part:
+                part = part.replace(trans, await translate(trans))
+    else:
+        part = await translate(part)
+    print(f"[{i}] translated")
+    return part
+
+async def translate_readme(readme) -> str:
+    parts = readme.split('\n## ')
+    print(f"{len(parts)} parts...")
+    parts = await asyncio.gather(
+        *[translate_part("## " + part, i) for i, part in enumerate(parts)]
+    )
+    return "\n\n".join(parts)
+
+with open("README.md", "r") as fp:
+    readme = fp.read()
+
+print("Translate readme...")
+readme = asyncio.run(translate_readme(readme))
+
+file = f"README-{iso}.md"
+with open(file, "w") as fp:
+    fp.write(readme)
+print(f'"{file}" saved')

g4f/Provider/AItianhu.py

@@ -71,21 +71,9 @@ class AItianhu(AsyncGeneratorProvider):
 
                     if "detail" not in line:
                         raise RuntimeError(f"Response: {line}")
-                    
-                    content = line["detail"]["choices"][0]["delta"].get("content")
+
+                    content = line["detail"]["choices"][0]["delta"].get(
+                        "content"
+                    )
                     if content:
                         yield content
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("proxy", "str"),
-            ("temperature", "float"),
-            ("top_p", "int"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"

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

g4f/Provider/ChatBase.py

@@ -58,15 +58,4 @@ class ChatBase(AsyncGeneratorProvider):
                     for incorrect_response in cls.list_incorrect_responses:
                         if incorrect_response in response_data:
                             raise RuntimeError("Incorrect response")
-                    yield stream.decode()
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                    yield stream.decode()

g4f/Provider/ChatForAi.py

@@ -57,16 +57,6 @@ class ChatForAi(AsyncGeneratorProvider):
                         raise RuntimeError(f"Response: {chunk.decode()}")
                     yield chunk.decode()
 
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
     
 def generate_signature(timestamp: int, message: str, id: str):
     buffer = f"{timestamp}:{id}:{message}:7YN8z6d6"

g4f/Provider/FreeGpt.py

@@ -47,16 +47,6 @@ class FreeGpt(AsyncGeneratorProvider):
                         raise RuntimeError("Rate limit reached")
                     yield chunk
 
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
     
 def generate_signature(timestamp: int, message: str, secret: str = ""):
     data = f"{timestamp}:{message}:{secret}"

g4f/Provider/GeekGpt.py

@@ -70,16 +70,4 @@ class GeekGpt(BaseProvider):
                     raise RuntimeError(f'error | {e} :', json_data)
                 
                 if content:
-                    yield content
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ('model', 'str'),
-            ('messages', 'list[dict[str, str]]'),
-            ('stream', 'bool'),
-            ('temperature', 'float'),
-        ]
-        param = ', '.join([': '.join(p) for p in params])
-        return f'g4f.provider.{cls.__name__} supports: ({param})'
+                    yield content

g4f/Provider/Liaobots.py

@@ -97,17 +97,3 @@ class Liaobots(AsyncGeneratorProvider):
                 async for stream in response.content.iter_any():
                     if stream:
                         yield stream.decode()
-
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("proxy", "str"),
-            ("auth", "str"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"

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));

g4f/Provider/Opchatgpts.py

@@ -56,16 +56,4 @@ class Opchatgpts(AsyncGeneratorProvider):
                         if line["type"] == "live":
                             yield line["data"]
                         elif line["type"] == "end":
-                            break
-
-    @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})"
+                            break

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

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>')) {

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

g4f/Provider/Ylokh.py

@@ -55,21 +55,4 @@ class Ylokh(AsyncGeneratorProvider):
                                 yield content
                 else:
                     chat = await response.json()
-                    yield chat["choices"][0]["message"].get("content")
-
-
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("proxy", "str"),
-            ("timeout", "int"),
-            ("temperature", "float"),
-            ("top_p", "float"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                    yield chat["choices"][0]["message"].get("content")

g4f/Provider/base_provider.py

@@ -3,6 +3,8 @@ from __future__ import annotations
 from asyncio import AbstractEventLoop
 from concurrent.futures import ThreadPoolExecutor
 from abc import ABC, abstractmethod
+from inspect import signature, Parameter
+from types import NoneType
 
 from .helper import get_event_loop, get_cookies, format_prompt
 from ..typing import CreateResult, AsyncResult, Messages
@@ -52,17 +54,42 @@ class BaseProvider(ABC):
             executor,
             create_func
         )
-
+    
     @classmethod
     @property
     def params(cls) -> str:
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+        if issubclass(cls, AsyncGeneratorProvider):
+            sig = signature(cls.create_async_generator)
+        elif issubclass(cls, AsyncProvider):
+            sig = signature(cls.create_async)
+        else:
+            sig = signature(cls.create_completion)
+
+        def get_type_name(annotation: type) -> str:
+            if hasattr(annotation, "__name__"):
+                annotation = annotation.__name__
+            elif isinstance(annotation, NoneType):
+                annotation = "None"
+            return str(annotation)
+        
+        args = "";
+        for name, param in sig.parameters.items():
+            if name in ("self", "kwargs"):
+                continue
+            if name == "stream" and not cls.supports_stream:
+                continue
+            if args:
+                args += ", "
+            args += "\n"
+            args += "    " + name
+            if name != "model" and param.annotation is not Parameter.empty:
+                args += f": {get_type_name(param.annotation)}"
+            if param.default == "":
+                args += ' = ""'
+            elif param.default is not Parameter.empty:
+                args += f" = {param.default}"
+        
+        return f"g4f.Provider.{cls.__name__} supports: ({args}\n)"
 
 
 class AsyncProvider(BaseProvider):

g4f/Provider/deprecated/Aibn.py

@@ -39,18 +39,6 @@ class Aibn(AsyncGeneratorProvider):
                 response.raise_for_status()
                 async for chunk in response.iter_content():
                     yield chunk.decode()
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
     
 
 def generate_signature(timestamp: int, message: str, secret: str = "undefined"):

g4f/Provider/deprecated/Ails.py

@@ -77,19 +77,6 @@ class Ails(AsyncGeneratorProvider):
                             yield token
 
 
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
-
-
 def _hash(json_data: dict[str, str]) -> SHA256:
     base_string: str = f'{json_data["t"]}:{json_data["m"]}:WI,2rU#_r:r~aF4aJ36[.Z(/8Rv93Rf:{len(json_data["m"])}'
 

g4f/Provider/deprecated/Aivvm.py

@@ -69,16 +69,4 @@ class Aivvm(BaseProvider):
             try:
                 yield chunk.decode("utf-8")
             except UnicodeDecodeError:
-                yield chunk.decode("unicode-escape")
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ('model', 'str'),
-            ('messages', 'list[dict[str, str]]'),
-            ('stream', 'bool'),
-            ('temperature', 'float'),
-        ]
-        param = ', '.join([': '.join(p) for p in params])
-        return f'g4f.provider.{cls.__name__} supports: ({param})'
+                yield chunk.decode("unicode-escape")

g4f/Provider/deprecated/ChatgptDuo.py

@@ -44,15 +44,4 @@ class ChatgptDuo(AsyncProvider):
 
     @classmethod
     def get_sources(cls):
-        return cls._sources
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+        return cls._sources

g4f/Provider/deprecated/CodeLinkAva.py

@@ -47,17 +47,4 @@ class CodeLinkAva(AsyncGeneratorProvider):
                             break
                         line = json.loads(line[6:-1])
                         if content := line["choices"][0]["delta"].get("content"):
-                            yield content
-
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                            yield content

g4f/Provider/deprecated/DfeHub.py

@@ -60,18 +60,3 @@ class DfeHub(BaseProvider):
             if b"content" in chunk:
                 data = json.loads(chunk.decode().split("data: ")[1])
                 yield (data["choices"][0]["delta"]["content"])
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-            ("presence_penalty", "int"),
-            ("frequency_penalty", "int"),
-            ("top_p", "int"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"

g4f/Provider/deprecated/EasyChat.py

@@ -87,21 +87,4 @@ class EasyChat(BaseProvider):
                     splitData = chunk.decode().split("data:")
 
                     if len(splitData) > 1:
-                        yield json.loads(splitData[1])["choices"][0]["delta"]["content"]
-
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-            ("presence_penalty", "int"),
-            ("frequency_penalty", "int"),
-            ("top_p", "int"),
-            ("active_server", "int"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                        yield json.loads(splitData[1])["choices"][0]["delta"]["content"]

g4f/Provider/deprecated/Equing.py

@@ -66,15 +66,4 @@ class Equing(BaseProvider):
                 if b'content' in line:
                     line_json = json.loads(line.decode('utf-8').split('data: ')[1])
                     if token := line_json['choices'][0]['delta'].get('content'):
-                        yield token
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                        yield token

g4f/Provider/deprecated/FastGpt.py

@@ -74,15 +74,4 @@ class FastGpt(BaseProvider):
                         ):
                             yield token
                 except:
-                    continue
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                    continue

g4f/Provider/deprecated/GetGpt.py

@@ -55,22 +55,6 @@ class GetGpt(BaseProvider):
                 line_json = json.loads(line.decode('utf-8').split('data: ')[1])
                 yield (line_json['choices'][0]['delta']['content'])
 
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ('model', 'str'),
-            ('messages', 'list[dict[str, str]]'),
-            ('stream', 'bool'),
-            ('temperature', 'float'),
-            ('presence_penalty', 'int'),
-            ('frequency_penalty', 'int'),
-            ('top_p', 'int'),
-            ('max_tokens', 'int'),
-        ]
-        param = ', '.join([': '.join(p) for p in params])
-        return f'g4f.provider.{cls.__name__} supports: ({param})'
-
 
 def _encrypt(e: str):
     t = os.urandom(8).hex().encode('utf-8')

g4f/Provider/deprecated/H2o.py

@@ -86,22 +86,4 @@ class H2o(AsyncGeneratorProvider):
                 f"{cls.url}/conversation/{conversationId}",
                 proxy=proxy,
             ) as response:
-                response.raise_for_status()
-
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-            ("truncate", "int"),
-            ("max_new_tokens", "int"),
-            ("do_sample", "bool"),
-            ("repetition_penalty", "float"),
-            ("return_full_text", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                response.raise_for_status()

g4f/Provider/deprecated/Lockchat.py

@@ -48,16 +48,4 @@ class Lockchat(BaseProvider):
             if b"content" in token:
                 token = json.loads(token.decode("utf-8").split("data: ")[1])
                 if token := token["choices"][0]["delta"].get("content"):
-                    yield (token)
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                    yield (token)

g4f/Provider/deprecated/Myshell.py

@@ -98,18 +98,6 @@ class Myshell(AsyncGeneratorProvider):
                         raise RuntimeError(f"Received unexpected message: {data_type}")
 
 
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
-
-
 def generate_timestamp() -> str:
     return str(
         int(

g4f/Provider/deprecated/V50.py

@@ -58,17 +58,4 @@ class V50(BaseProvider):
         )
 
         if "https://fk1.v50.ltd" not in response.text:
-            yield response.text
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("temperature", "float"),
-            ("top_p", "int"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+            yield response.text

g4f/Provider/deprecated/Vitalentum.py

@@ -50,18 +50,4 @@ class Vitalentum(AsyncGeneratorProvider):
                             break
                         line = json.loads(line[6:-1])
                         if content := line["choices"][0]["delta"].get("content"):
-                            yield content
-
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool"),
-            ("proxy", "str"),
-            ("temperature", "float"),
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+                            yield content

g4f/Provider/deprecated/Wuguokai.py

@@ -54,15 +54,4 @@ class Wuguokai(BaseProvider):
         if len(_split) > 1:
             yield _split[1].strip()
         else:
-            yield _split[0].strip()
-
-    @classmethod
-    @property
-    def params(cls):
-        params = [
-            ("model", "str"),
-            ("messages", "list[dict[str, str]]"),
-            ("stream", "bool")
-        ]
-        param = ", ".join([": ".join(p) for p in params])
-        return f"g4f.provider.{cls.__name__} supports: ({param})"
+            yield _split[0].strip()

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)

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")))

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()

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})"

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,

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']")))

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})"

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));

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()

g4f/__init__.py

@@ -1,8 +1,8 @@
 from __future__ import annotations
 from requests   import get
 from .models    import Model, ModelUtils, _all_models
-from .Provider  import BaseProvider, RetryProvider
-from .typing    import Messages, CreateResult, Union, List
+from .Provider  import BaseProvider, AsyncGeneratorProvider, RetryProvider
+from .typing    import Messages, CreateResult, AsyncResult, Union, List
 from .          import debug
 
 version       = '0.1.8.7'
@@ -80,13 +80,15 @@ class ChatCompletion:
                            messages : Messages,
                            provider : Union[type[BaseProvider], None] = None,
                            stream   : bool = False,
-                           ignored  : List[str] = None, **kwargs) -> str:
-        
-        if stream:
-            raise ValueError('"create_async" does not support "stream" argument')
-
+                           ignored  : List[str] = None,
+                           **kwargs) -> Union[AsyncResult, str]:
         model, provider = get_model_and_provider(model, provider, False, ignored)
 
+        if stream:
+            if isinstance(provider, type) and issubclass(provider, AsyncGeneratorProvider):
+                return await provider.create_async_generator(model.name, messages, **kwargs)
+            raise ValueError(f'{provider.__name__} does not support "stream" argument')
+
         return await provider.create_async(model.name, messages, **kwargs)
 
 class Completion:

requirements.txt

@@ -1,6 +1,6 @@
 requests
 pycryptodome
-curl_cffi
+curl_cffi>=0.5.10b4
 aiohttp
 certifi
 browser_cookie3
@@ -23,6 +23,6 @@ flask
 py-arkose-generator
 asyncstdlib
 async-property
-selenium
+undetected-chromedriver
 asyncstdlib
-async_property
+async_property