python run.py

Enter any words but reply nothing, terminal says
list index out of range

i did it and it says this
Running on port 1337

• Serving Flask app 'server.app'
• Debug mode: off
  WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
• Running on all addresses (0.0.0.0)
  and then my ip and stuff but how do iuse the clone

Initially, its build over the top of gpt4free using some Rev APIs"s to make it free then why are you @xtekky using Open AI Key in Backend it means you are routing from your API Key.

gpt_resp = post('https://api.openai.com/v1/chat/completions', headers = {'Authorization': 'Bearer %s' % self.openai_key}, json = { 'model' : request.json['model'], 'messages' : conversation, 'stream' : True}, stream = True)

As mentioned in the readme, a theme picker addition is required for changing the chat themes.

{"_action":"_ask","error":"an error occurred string indices must be integers","success":false}

I get this error when I try to run the program
self.openai_key = os.environ["OPENAI_API_KEY"] or config['openai_key']

A page or script is accessing at least one of navigator.userAgent, navigator.appVersion, and navigator.platform. Starting in Chrome 101, the amount of information available in the User Agent string will be reduced.
To fix this issue, replace the usage of navigator.userAgent, navigator.appVersion, and navigator.platform with feature detection, progressive enhancement, or migrate to navigator.userAgentData.
Note that for performance reasons, only the first access to one of the properties is shown.
how to solve this?thank you

conversions

• double confirm when delete
• support output

theme

easy switch black to white.

Currently the ui always scroll to the bottom when text generation takes place. It would be nice to allow scrolling up to start reading answers while generating

i saw this project of chat-gpt integrated with ElevenLabs api to generate a text-to-speech button and think that is an awesome idea, any chance to implement something similar?
https://github.com/cogentapps/chat-with-gpt

My device screen size is 17 inches and a black bar is visible at the bottom of the screen

I have fixed this issue.

Kindly accept my PR.

So SLOWLY that I don't want to use it.

please...

Hi,

Currently, user preferences (like model preference) are not persisted.
I find myself all the time changing it back to my preferred model.

We could store the user's preference on localStorage.

This message:

is removed when pressing

The Error in the CLI is ....

The opened Port Results are....

Platform Windows

{"_action":"_ask","error":"an error occurred HTTPSConnectionPool(host='api.openai.com', port=443): Max retries exceeded with url: /v1/chat/completions (Caused by SSLError(SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1131)')))","success":false}

I don't konw why i sent information, it sents"{ "error": { "message": "You exceeded your current quota, please check your plan and billing details.", "type": "insufficient_quota", "param": null, "code": null } }"；Can you tell me what happend, thank you so much!!!

Hi

its is possible to have the ability to custom the OpenAi API call with those parameters from UI.so we can get granulated response results

Tokens
Temperature
Top_P
Presence Penalty
Frequency Penalty

regards

Is there a way to bypass the token limit ? I get pretty much limited after 3-4 requests

when executing the command python run.py
gives an error message :
Traceback (most recent call last):
File "C:\chatgpt-clone\run.py", line 19, in
backend_api = Backend_Api(app, config)
^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\chatgpt-clone\server\backend.py", line 17, in init
self.openai_key = os.environ["OPENAI_API_KEY"] or config['openai_key']
~~~~~~~~~~^^^^^^^^^^^^^^^^^^
File "", line 679, in getitem
KeyError: 'OPENAI_API_KEY'

What should I do ?
openai_key I entered in the file config.json

Hi

can we have back the original black them as default for theme picker feature .

regards

Is there a way to integrate with autogpt? It needs OpenAI api link. (Project link: https://github.com/Significant-Gravitas/Auto-GPT)

جميل جدا

what should i do for this error:

{"_action":"_ask","error":"an error occurred HTTPSConnectionPool(host='ddg-api.herokuapp.com', port=443): Max retries exceeded with url: /search?query=hi&limit=3 (Caused by ConnectTimeoutError(<urllib3.connection.HTTPSConnection object at 0x0000027A0A0FD210>, 'Connection to ddg-api.herokuapp.com timed out. (connect timeout=None)'))","success":false}

selfhosted version
after send message to bot i see in cli message(screenshot)
and bot don't give answer

I got the following error message

{ "error": { "message": "You exceeded your current quota, please check your plan and billing details.", "type": "insufficient_quota", "param": null, "code": null } }

I ALREADY LAUNCHED THE GPT BUT I CANT USE IT, HOW CAN I CHANGE THE API? T.Y

WHEN I TRY TO INSTALL USING: pip install -r requirements.txt

Fatal error in launcher: "Unable to create process using '"C:\Users\flavi\chatgpt-clone\venv\Scripts\python.exe" "C:\WINDOWS\system32\chatgpt-clone\venv\Scripts\pip.exe" install -r requirements.txt': O sistema nÒo pode encontrar o arquivo especificado."

I want to suggest adding a regenerate button. It will help the user to quickly regenerate the result if the last output is not as per the expectation. This feature is not related to any existing issues or pull requests.

hi

can we have the ability to upload documents to a milvus DB (vector DB) to enhance the bots memories and responses results .

regards

After using docker compose up i received following output

chatgpt-clone  | Running on port 1338
chatgpt-clone  |  * Serving Flask app 'server.app'
chatgpt-clone  |  * Debug mode: off
chatgpt-clone  | WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
chatgpt-clone  |  * Running on all addresses (0.0.0.0)
chatgpt-clone  |  * Running on http://127.0.0.1:1338
chatgpt-clone  |  * Running on http://172.18.0.2:1338
chatgpt-clone  | Press CTRL+C to quit

but in all adresses in my browser I get a blank page with message: ERR_CONNECTION_REFUSED
Thank for any help.

what is wrong is it ?

Is there any way to implement file directory access into this when running locally? I am interested in providing the bot with context and pre-existing files to use as a framework.

https://github.com/mayooear/gpt4-pdf-chatbot-langchain enables reading from PDF files, however I wonder if it's possible to provide it with a directory consisting of .txt, .docx, .py, etc files that it can then use as a pre-existing context

{"_action":"_ask","error":"an error occurred string indices must be integers, not 'str'","success":false}

Color picker does not show the color name on hover the radio button. Which makes it harder to select the color.

There should be a title for each radio button, when user hover over it.

I would love to raise a PR for this.

Hello, is it possible or do you agree if we implement multiple bot engine in this project?
Recently I got API quota limit, so maybe it's better to transparently show which API/ bot provider we use in backend and show the remaining quota per bot and we can manually choose it.
For the user interface, maybe it looks like poe

I run with docker. Mac M1, I am seeing this error when response form chatgpt it long.

chatgpt-clone  | Error on request:
chatgpt-clone  | Traceback (most recent call last):
chatgpt-clone  |   File "/root/.local/lib/python3.8/site-packages/urllib3/response.py", line 761, in _update_chunk_length
chatgpt-clone  |     self.chunk_left = int(line, 16)
chatgpt-clone  | ValueError: invalid literal for int() with base 16: b''
chatgpt-clone  | 
chatgpt-clone  | During handling of the above exception, another exception occurred:
chatgpt-clone  | 
chatgpt-clone  | Traceback (most recent call last):
chatgpt-clone  |   File "/root/.local/lib/python3.8/site-packages/urllib3/response.py", line 444, in _error_catcher
chatgpt-clone  |     yield
chatgpt-clone  |   File "/root/.local/lib/python3.8/site-packages/urllib3/response.py", line 828, in read_chunked
chatgpt-clone  |     self._update_chunk_length()
chatgpt-clone  |   File "/root/.local/lib/python3.8/site-packages/urllib3/response.py", line 765, in _update_chunk_length
chatgpt-clone  |     raise InvalidChunkLength(self, line)
chatgpt-clone  | urllib3.exceptions.InvalidChunkLength: InvalidChunkLength(got length b'', 0 bytes read)

https://chatforai.com/
https://chatbot.theb.ai/#/chat

GPT 3.5
https://chat9.yqcloud.top/#/chat/
https://ai.usesless.com/chat/
https://xc.com/
https://dev.yqcloud.top
https://chat.gptalk.net/
https://chat.yqcloud.top/#

Feature requests:

1. As someone that is impatient, I would like to be able to scroll while ChatGPT is busy typing the next results.
2. As someone with ADHD, I would like to already write and pass on my next prompt.
3. As someone that spends too many hours staring at screens, I would like to have the option to have ChatGPT's response appear instantly after a (...loading.. ) screen.

I don't understand how, but many thanks for your gift to this world. May open-source always win.