haseeb-heaven / code-interpreter

An innovative open-source Code Interpreter with (GPT,Gemini,Claude,LLaMa) models.

Hey, I'm Heaven 👋

I'm a coder and developer who loves to explore new technologies and create awesome things with them. I have expertise in tech and gaming industry, and I work on OpenAI tools, generative AI, API & SDK integration, and game development projects.

👨‍💻 What I do

• I’m currently learning AI tech, deep learning, and AI models.
• I’m looking to collaborate on cool and open-source projects.
• I write about tech related topics on my blogs and YouTube channel.
• I have worked in AAA gaming industry and have created security tools to enhance gaming experience.

Programming Languages

Top Repositories

📫 How to reach me

code-interpreter's Issues

No more palm :(

Can you please update the code as there is no more palm api access :(

Integration with Naas Chat (ChatGPT clone)

Is your feature request related to a problem? Please describe.
Not related to a problem.

Describe the solution you'd like
I want to discuss a potential integration in our chatGPT clone; Naas Chat.

Open for a chat?

Local-model not working, as it expects openai api key.

Hello!
Regarding this issue, I am currently using LM Studio. When using local-model, it does not work. As I can see in code, in interpreter_lib.py, line 324, variable custom_llm_provider is set to 'openai', so it expects de openai api key. Which has to be the value of this variable when using open-source LLMs as Mistral?

Support for Gemini 1.5 Pro

It supports a token count of 1 million+ and It supports both Images and Videos as well

Model: gemini-1.5-pro-Iatest

It is in private preview though, so not everyone has access to it yet.

Why not use LiteLLM?

Hi @haseeb-heaven,

Curious what litellm was missing to be useful here? Any feedback would be great
https://github.com/haseeb-heaven/open-code-interpreter/blob/4c142036c430b1184c8eb420c8af804638ee73d1/interpreter.py#L7

Offline-Model via LM Studio does not work as expected / local-model confusion

Hi,
the readme says, this project would work with a local LM Studio server. The docs says, you have to edit the config/offline-model.config, which does not exist, but the file with the similar purpose is 'config/local-model.config', at least I guess it's meant like that. When trying to run 'python interpreter.py -md 'code' -m 'local-model' -dc' it will ask for the .env with keys for the remote LMs. Creating an empty .env or and .env with empty keys does not change anything. Looks like a bug to me.
Would be great to use this project without using proprietary models.
Thanks, BR Tobias

Open-Code-Interpreter can't chat with my data

this is almost perfect, but I can only see the code the llm drops, I want to also chat with it, for debugging purposes, don't get me wrong, I want to execute code, but I want to ask questions, and do other things.

GPT 3.5 Turbo model not working

We are currently experiencing a temporary downtime with this model, as we work diligently to address and resolve the issue. During this time, we are exploring alternative service providers who can offer reliable solutions to ensure uninterrupted access to our services.

Currently working providers for GPT 3.5 Turbo which works on server.

• FreeGpt
• Aivvm
• Vercel

Package not available 'Readline' - Windows

Kindly guide. (base) python -m 'code-llama' -md 'code'
Traceback (most recent call last):

  File "C:\Users\test\open-code-interpreter\[http://interpreter.py](https://t.co/Ag94dpbB78)", line 14, in <module>
    from libs.utility_manager import UtilityManager
  File "C:\Users\test\open-code-interpreter\libs\utility_manager.py", line 3, in <module>
    import readline
ModuleNotFoundError: No module named 'readline'

Error: File "", line 1 SyntaxError: invalid syntax

hi im getting problems running this using claude or groq models error i get is

install django
1 sudo apt-get update

Error: File "", line 1
sudo apt-get update
^^^
SyntaxError: invalid syntax