Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

When use ollama with model llama3:70b, the code cannot run #1220

Open
neotelsha opened this issue Apr 22, 2024 · 11 comments
Open

When use ollama with model llama3:70b, the code cannot run #1220

neotelsha opened this issue Apr 22, 2024 · 11 comments

Comments

@neotelsha
Copy link

Describe the bug

when use ollama with model llama3:70b, the python code return with a "`" ,caused the code cannot be executed.

Reproduce

  1. run interpreter by this command . interpreter --model ollama/llama3:70b
  2. input please ananlyze d:\test\dumpstack.log

Expected behavior

it should give me some output

Screenshots

error
error2

Open Interpreter version

0.2.0

Python version

3.12.2

Operating System name and version

windows 11

Additional context

No response

@jkaldon
Copy link

jkaldon commented Apr 23, 2024

I'm also having this problem.

@bluekeybo
Copy link

same here, with the llama3 8b model. I do get the backtick and also get random responses that don't solve the issue. Sometimes it responds to previous questions even though I ask for something else.

@Supercompact
Copy link

Supercompact commented Apr 23, 2024

Same issue running llama3-8b on MacOS 14.4 using LM Studio.

@MnLgt
Copy link

MnLgt commented Apr 24, 2024

Same here; ollama/llama3-8b on MacOS 13.5.2

@jasonkneen
Copy link

same -- goes very mad

@Notnaton
Copy link
Collaborator

Update to version 0.2.5

pip install --force-reinstall --upgrade open-interpreter

@MnLgt
Copy link

MnLgt commented Apr 26, 2024

Update to version 0.2.5

pip install --force-reinstall --upgrade open-interpreter

Seems to be working now after the reinstall. Thanks!

@jasonkneen
Copy link

Tried it and installed most got errors:

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
gensim 4.3.0 requires FuzzyTM>=0.4.0, which is not installed.
streamlit 1.30.0 requires packaging<24,>=16.8, but you have packaging 24.0 which is incompatible.
mdit-py-plugins 0.3.0 requires markdown-it-py<3.0.0,>=1.0.0, but you have markdown-it-py 3.0.0 which is incompatible.
faster-whisper 0.10.0 requires tokenizers<0.16,>=0.13, but you have tokenizers 0.19.1 which is incompatible.
botocore 1.31.64 requires urllib3<2.1,>=1.25.4; python_version >= "3.10", but you have urllib3 2.2.1 which is incompatible.
python-lsp-server 1.7.2 requires jedi<0.19.0,>=0.17.2, but you have jedi 0.19.1 which is incompatible.
conda-repo-cli 1.0.75 requires clyent==1.2.1, but you have clyent 1.2.2 which is incompatible.
conda-repo-cli 1.0.75 requires python-dateutil==2.8.2, but you have python-dateutil 2.9.0.post0 which is incompatible.
s3fs 2023.10.0 requires fsspec==2023.10.0, but you have fsspec 2024.3.1 which is incompatible.
transformers 4.39.3 requires tokenizers<0.19,>=0.14, but you have tokenizers 0.19.1 which is incompatible.
spyder 5.4.3 requires jedi<0.19.0,>=0.17.2, but you have jedi 0.19.1 which is incompatible.
ane-transformers 0.1.1 requires protobuf<=3.20.1,>=3.1.0, but you have protobuf 3.20.3 which is incompatible.

Will it work? lol

Also still getting this:

/opt/anaconda3/lib/python3.11/site-packages/pydantic/_internal/_fields.py:160: UserWarning: Field "model_name" has conflict with protected namespace "model_".

You may be able to resolve this warning by setting `model_config['protected_namespaces'] = ()`.
  warnings.warn(
/opt/anaconda3/lib/python3.11/site-packages/pydantic/_internal/_fields.py:160: UserWarning: Field "model_info" has conflict with protected namespace "model_".

You may be able to resolve this warning by setting `model_config['protected_namespaces'] = ()`.
  warnings.warn(

(although got this before and still works)

@jasonkneen
Copy link

Re #1220 (comment)

So when I try with various models I get odd things happening where it's almost telling me it's thoughts or just coming back with odd stuff:

> ls

  Plan Recap: The user needs help setting up a repository. We previously created a new directory and initialized it as a
  Git repository. Now, we will list the contents of the directory to verify that everything is set up correctly. Current
  State: We are now listing the contents of the directory.



  `
  ls


  ---------------------------------------------------------------------------
  NameError                                 Traceback (most recent call last)
  Cell In[3], line 2
        1 print('##active_line1##')
  ----> 2 ls

  NameError: name 'ls' is not defined

> ##ls

  Plan Recap: The user wants to list the contents of the current directory. We will use theos module to execute thels
  command. Current State: We are now listing the contents of the directory.



  `
  import os
  print(os.listdir())


  ['awesome-openai-vision-api-experiments', 'gpt-video', 'HeyGPT-AI-Voice-Companion', 'personal-ai', '.DS_Store',
  'gpt-pilot-backup-0-1-10-14a2d3ef', 'vision.py', 'OpenVoice', 'macOSpilot-ai-assistant', '__pycache__', 'whisper.cpp',
  'oobabooga', 'MagicLLight', 'rem', '01', 'gpt-pilot', 'bot', 'testvision.py', '.vscode',
  'LM-Studio-Voice-Conversation', '~']

@jasonkneen
Copy link

latest I have now is it always responding with "a plan" so

> hello there
                                                                                                                                                                                                                                                                                                                 
  Hello, Open Interpreter here. I'll follow the plan as described:                                                                                                                                                                                                                                               
                                                                                                                                                                                                                                                                                                                 
   1 Greet the user.                                                                                                                                                                                                                                                                                             
                                                                                                                                                                                                                                                                                                                 
  Plan: Send a message to greet the user.                                                                                                                                                                                                                                                                        
                                                                                                                                                                                                                                                                                                                 
                                                                                                                                                                                                                                                                                                                 
   print("Hello there!                                                                                                                                                                                                                                                                                           
                                                                                                                                                                                                                                                                                                                 
                                                                                                                                                                                                                                                                                                                 
> how are you today?
                                                                                                                                                                                                                                                                                                                 
  I'm just a program, Open Interpreter, but I'll do my best to help you out today! Let's get started with your task.                                                                                                                                                                                             
                                                                                                                                                                                                                                                                                                                 
  Plan: Get the user's current date and time for reference.                                                                                                                                                                                                                                                      
                                                                                                                                                                                                                                                                                                                 
> 

I did an update to 0.25 but I think there's been 2 updates to that so how do I know I've got the right one?

@oloboy
Copy link

oloboy commented Apr 30, 2024

Same issue running llama3 on windows
but the update to 0.2.5 solve the problem
pip install --force-reinstall --upgrade open-interpreter

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

8 participants