> By using this repository or any code related to it, you agree to the [legal notice](LEGAL_NOTICE.md). The author is not responsible for any copies, forks, re-uploads made by other users, or anything else related to GPT4Free. This is the author's only account and repository. To prevent impersonation or irresponsible actions, please comply with the GNU GPL license this Repository uses.
> [!Note]
> Latest pypi version: [`0.1.8.4`](https://pypi.org/project/g4f/0.1.8.4)
```sh
pip install -U g4f
```
## π What's New
- Join our Telegram Channel: [t.me/g4f_channel](https://telegram.me/g4f_channel)
- Join our Discord Group: [discord.gg/XfybzPXPH5](https://discord.gg/XfybzPXPH5)
- Explore the g4f Documentation (unfinished): [g4f.mintlify.app](https://g4f.mintlify.app) | Contribute to the docs via: [github.com/xtekky/gpt4free-docs](https://github.com/xtekky/gpt4free-docs)
## π Table of Contents
- [π What's New](#-whats-new)
- [π Table of Contents](#-table-of-contents)
- [π οΈ Getting Started](#οΈ-getting-started)
- [Prerequisites:](#prerequisites)
- [Setting up the project:](#setting-up-the-project)
- [Install using pypi](#install-using-pypi)
- [or](#or)
- [Setting up with Docker:](#setting-up-with-docker)
- [π‘ Usage](#-usage)
- [The `g4f` Package](#the-g4f-package)
- [ChatCompletion](#chatcompletion)
- [Completion](#completion)
- [Providers](#providers)
- [Cookies Required](#cookies-required)
- [Async Support](#async-support)
- [Proxy and Timeout Support](#proxy-and-timeout-support)
- [Interference openai-proxy API (Use with openai python package)](#interference-openai-proxy-api-use-with-openai-python-package)
- [Run interference API from PyPi package](#run-interference-api-from-pypi-package)
- [Run interference API from repo](#run-interference-api-from-repo)
- [π Providers and Models](#-providers-and-models)
- [GPT-4](#gpt-4)
- [GPT-3.5](#gpt-35)
- [Other](#other)
- [Models](#models)
- [π Related GPT4Free Projects](#-related-gpt4free-projects)
- [π€ Contribute](#-contribute)
- [Create Provider with AI Tool](#create-provider-with-ai-tool)
- [Create Provider](#create-provider)
- [π Contributors](#-contributors)
- [Β©οΈ Copyright](#οΈ-copyright)
- [β Star History](#-star-history)
- [π License](#-license)
## π οΈ Getting Started
#### Prerequisites:
1. [Download and install Python](https://www.python.org/downloads/) (Version 3.10+ is recommended).
#### Setting up the project:
##### Install using pypi
```
pip install -U g4f
```
##### or
1. Clone the GitHub repository:
```
git clone https://github.com/xtekky/gpt4free.git
```
2. Navigate to the project directory:
```
cd gpt4free
```
3. (Recommended) Create a Python virtual environment:
You can follow the [Python official documentation](https://docs.python.org/3/tutorial/venv.html) for virtual environments.
```
python3 -m venv venv
```
4. Activate the virtual environment:
- On Windows:
```
.\venv\Scripts\activate
```
- On macOS and Linux:
```
source venv/bin/activate
```
5. Install the required Python packages from `requirements.txt`:
```
pip install -r requirements.txt
```
6. Create a `test.py` file in the root folder and start using the repo, further Instructions are below
```py
import g4f
...
```
##### Setting up with Docker:
If you have Docker installed, you can easily set up and run the project without manually installing dependencies.
1. First, ensure you have both Docker and Docker Compose installed.
- [Install Docker](https://docs.docker.com/get-docker/)
- [Install Docker Compose](https://docs.docker.com/compose/install/)
2. Clone the GitHub repo:
```bash
git clone https://github.com/xtekky/gpt4free.git
```
3. Navigate to the project directory:
```bash
cd gpt4free
```
4. Build the Docker image:
```bash
docker-compose build
```
5. Start the service using Docker Compose:
```bash
docker-compose up
```
Your server will now be running at `http://localhost:1337`. You can interact with the API or run your tests as you would normally.
To stop the Docker containers, simply run:
```bash
docker-compose down
```
> [!Note]
> When using Docker, any changes you make to your local files will be reflected in the Docker container thanks to the volume mapping in the `docker-compose.yml` file. If you add or remove dependencies, however, you'll need to rebuild the Docker image using `docker-compose build`.
## π‘ Usage
### The `g4f` Package
#### ChatCompletion
```python
import g4f
g4f.debug.logging = True # Enable logging
g4f.check_version = False # Disable automatic version checking
print(g4f.version) # Check version
print(g4f.Provider.Ails.params) # Supported args
# Automatic selection of provider
# Streamed completion
response = g4f.ChatCompletion.create(
model="gpt-3.5-turbo",
messages=[{"role": "user", "content": "Hello"}],
stream=True,
)
for message in response:
print(message, flush=True, end='')
# Normal response
response = g4f.ChatCompletion.create(
model=g4f.models.gpt_4,
messages=[{"role": "user", "content": "Hello"}],
) # Alternative model setting
print(response)
```
##### Completion
```python
import g4f
allowed_models = [
'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='say this is a test'
)
print(response)
```
##### Providers
```python
import g4f
from g4f.Provider import (
AItianhu,
Aichat,
Bard,
Bing,
ChatBase,
ChatgptAi,
OpenaiChat,
Vercel,
You,
Yqcloud,
)
# Set with provider
response = g4f.ChatCompletion.create(
model="gpt-3.5-turbo",
provider=g4f.Provider.Aichat,
messages=[{"role": "user", "content": "Hello"}],
stream=True,
)
for message in response:
print(message)
```
##### 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:
```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": "Suggest me a name."}],
browser=browser
)
print(f"{idx}:", response)
browser.quit()
```
##### Cookies Required
Cookies are essential for the proper functioning of some service providers. It is imperative to maintain an active session, typically achieved by logging into your account.
When running the g4f package locally, the package automatically retrieves cookies from your web browser using the `get_cookies` function. However, if you're not running it locally, you'll need to provide the cookies manually by passing them as parameters using the `cookies` parameter.
```python
import g4f
from g4f.Provider import (
Bing,
HuggingChat,
OpenAssistant,
)
# Usage
response = g4f.ChatCompletion.create(
model=g4f.models.default,
messages=[{"role": "user", "content": "Hello"}],
provider=Bing,
#cookies=g4f.get_cookies(".google.com"),
cookies={"cookie_name": "value", "cookie_name2": "value2"},
auth=True
)
```
##### Async Support
To enhance speed and overall performance, execute providers asynchronously. The total execution time will be determined by the duration of the slowest provider's execution.
```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": "Hello"}],
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())
```
##### Proxy and Timeout Support
All providers support specifying a proxy and increasing timeout in the create functions.
```python
import g4f
response = g4f.ChatCompletion.create(
model=g4f.models.default,
messages=[{"role": "user", "content": "Hello"}],
proxy="http://host:port",
# or socks5://user:pass@host:port
timeout=120, # in secs
)
print(f"Result:", response)
```
### Interference openai-proxy API (Use with openai python package)
#### Run interference API from PyPi package
```python
from g4f.api import run_api
run_api()
```
#### Run interference API from repo
If you want to use the embedding function, you need to get a Hugging Face token. You can get one at [Hugging Face Tokens](https://huggingface.co/settings/tokens). Make sure your role is set to write. If you have your token, just use it instead of the OpenAI api-key.
Run server:
```sh
g4f api
```
or
```sh
python -m g4f.api
```
```python
import openai
# Set your Hugging Face token as the API key if you use embeddings
# If you don't use embeddings, leave it empty
openai.api_key = "YOUR_HUGGING_FACE_TOKEN" # Replace with your actual token
# Set the API base URL if needed, e.g., for a local development environment
openai.api_base = "http://localhost:1337/v1"
def main():
chat_completion = openai.ChatCompletion.create(
model="gpt-3.5-turbo",
messages=[{"role": "user", "content": "write a poem about a tree"}],
stream=True,
)
if isinstance(chat_completion, dict):
# Not streaming
print(chat_completion.choices[0].message.content)
else:
# Streaming
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()
```
## π Providers and Models
### GPT-4
| Website | Provider | GPT-3.5 | GPT-4 | Stream | Status | Auth |
| ------ | ------- | ------- | ----- | ------ | ------ | ---- |
| [bing.com](https://bing.com/chat) | `g4f.Provider.Bing` | β | βοΈ | βοΈ |  | β |
| [chat.geekgpt.org](https://chat.geekgpt.org) | `g4f.Provider.GeekGpt` | βοΈ | βοΈ | βοΈ |  | β |
| [gptchatly.com](https://gptchatly.com) | `g4f.Provider.GptChatly` | βοΈ | βοΈ | β |  | β |
| [liaobots.site](https://liaobots.site) | `g4f.Provider.Liaobots` | βοΈ | βοΈ | βοΈ |  | β |
| [www.phind.com](https://www.phind.com) | `g4f.Provider.Phind` | β | βοΈ | βοΈ |  | β |
| [raycast.com](https://raycast.com) | `g4f.Provider.Raycast` | βοΈ | βοΈ | βοΈ |  | βοΈ |
### GPT-3.5
| Website | Provider | GPT-3.5 | GPT-4 | Stream | Status | Auth |
| ------ | ------- | ------- | ----- | ------ | ------ | ---- |
| [www.aitianhu.com](https://www.aitianhu.com) | `g4f.Provider.AItianhu` | βοΈ | β | βοΈ |  | β |
| [chat3.aiyunos.top](https://chat3.aiyunos.top/) | `g4f.Provider.AItianhuSpace` | βοΈ | β | βοΈ |  | β |
| [e.aiask.me](https://e.aiask.me) | `g4f.Provider.AiAsk` | βοΈ | β | βοΈ |  | β |
| [chat-gpt.org](https://chat-gpt.org/chat) | `g4f.Provider.Aichat` | βοΈ | β | β |  | β |
| [www.chatbase.co](https://www.chatbase.co) | `g4f.Provider.ChatBase` | βοΈ | β | βοΈ |  | β |
| [chatforai.store](https://chatforai.store) | `g4f.Provider.ChatForAi` | βοΈ | β | βοΈ |  | β |
| [chatgpt.ai](https://chatgpt.ai) | `g4f.Provider.ChatgptAi` | βοΈ | β | βοΈ |  | β |
| [chatgptx.de](https://chatgptx.de) | `g4f.Provider.ChatgptX` | βοΈ | β | βοΈ |  | β |
| [chat-shared2.zhile.io](https://chat-shared2.zhile.io) | `g4f.Provider.FakeGpt` | βοΈ | β | βοΈ |  | β |
| [freegpts1.aifree.site](https://freegpts1.aifree.site/) | `g4f.Provider.FreeGpt` | βοΈ | β | βοΈ |  | β |
| [gptalk.net](https://gptalk.net) | `g4f.Provider.GPTalk` | βοΈ | β | βοΈ |  | β |
| [ai18.gptforlove.com](https://ai18.gptforlove.com) | `g4f.Provider.GptForLove` | βοΈ | β | βοΈ |  | β |
| [gptgo.ai](https://gptgo.ai) | `g4f.Provider.GptGo` | βοΈ | β | βοΈ |  | β |
| [hashnode.com](https://hashnode.com) | `g4f.Provider.Hashnode` | βοΈ | β | βοΈ |  | β |
| [app.myshell.ai](https://app.myshell.ai/chat) | `g4f.Provider.MyShell` | βοΈ | β | βοΈ |  | β |
| [noowai.com](https://noowai.com) | `g4f.Provider.NoowAi` | βοΈ | β | βοΈ |  | β |
| [chat.openai.com](https://chat.openai.com) | `g4f.Provider.OpenaiChat` | βοΈ | β | βοΈ |  | βοΈ |
| [theb.ai](https://theb.ai) | `g4f.Provider.Theb` | βοΈ | β | βοΈ |  | βοΈ |
| [sdk.vercel.ai](https://sdk.vercel.ai) | `g4f.Provider.Vercel` | βοΈ | β | βοΈ |  | β |
| [you.com](https://you.com) | `g4f.Provider.You` | βοΈ | β | βοΈ |  | β |
| [chat9.yqcloud.top](https://chat9.yqcloud.top/) | `g4f.Provider.Yqcloud` | βοΈ | β | βοΈ |  | β |
| [chat.acytoo.com](https://chat.acytoo.com) | `g4f.Provider.Acytoo` | βοΈ | β | βοΈ |  | β |
| [aibn.cc](https://aibn.cc) | `g4f.Provider.Aibn` | βοΈ | β | βοΈ |  | β |
| [ai.ls](https://ai.ls) | `g4f.Provider.Ails` | βοΈ | β | βοΈ |  | β |
| [chatgpt4online.org](https://chatgpt4online.org) | `g4f.Provider.Chatgpt4Online` | βοΈ | β | βοΈ |  | β |
| [chat.chatgptdemo.net](https://chat.chatgptdemo.net) | `g4f.Provider.ChatgptDemo` | βοΈ | β | βοΈ |  | β |
| [chatgptduo.com](https://chatgptduo.com) | `g4f.Provider.ChatgptDuo` | βοΈ | β | β |  | β |
| [chatgptfree.ai](https://chatgptfree.ai) | `g4f.Provider.ChatgptFree` | βοΈ | β | β |  | β |
| [chatgptlogin.ai](https://chatgptlogin.ai) | `g4f.Provider.ChatgptLogin` | βοΈ | β | βοΈ |  | β |
| [cromicle.top](https://cromicle.top) | `g4f.Provider.Cromicle` | βοΈ | β | βοΈ |  | β |
| [gptgod.site](https://gptgod.site) | `g4f.Provider.GptGod` | βοΈ | β | βοΈ |  | β |
| [opchatgpts.net](https://opchatgpts.net) | `g4f.Provider.Opchatgpts` | βοΈ | β | βοΈ |  | β |
| [chat.ylokh.xyz](https://chat.ylokh.xyz) | `g4f.Provider.Ylokh` | βοΈ | β | βοΈ |  | β |
### Other
| Website | Provider | GPT-3.5 | GPT-4 | Stream | Status | Auth |
| ------ | ------- | ------- | ----- | ------ | ------ | ---- |
| [bard.google.com](https://bard.google.com) | `g4f.Provider.Bard` | β | β | β |  | βοΈ |
| [deepinfra.com](https://deepinfra.com) | `g4f.Provider.DeepInfra` | β | β | βοΈ |  | β |
| [huggingface.co](https://huggingface.co/chat) | `g4f.Provider.HuggingChat` | β | β | βοΈ |  | βοΈ |
| [www.llama2.ai](https://www.llama2.ai) | `g4f.Provider.Llama2` | β | β | βοΈ |  | β |
| [open-assistant.io](https://open-assistant.io/chat) | `g4f.Provider.OpenAssistant` | β | β | βοΈ |  | βοΈ |
### Models
| 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/) |
## π Related GPT4Free Projects
