github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #838] how to use ollama with open-interpreter? #402

New Issue
Closed
opened 2026-04-12 10:02:57 -05:00 by GiteaMirror · 17 comments
GiteaMirror commented 2026-04-12 10:02:57 -05:00
Owner
Copy Link

Originally created by @wuyongyi on GitHub (Oct 18, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/838

I noticed that open-interpreter utilizes litellm to communicate with llms. While litellm can utilise ollama as a backend to respond to prompts, I have been unable to find a way to utilise ollama within open-interpreter. Does anyone have any experience or knowledge regarding this?

Originally created by @wuyongyi on GitHub (Oct 18, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/838 I noticed that open-interpreter utilizes litellm to communicate with llms. While litellm can utilise ollama as a backend to respond to prompts, I have been unable to find a way to utilise ollama within open-interpreter. Does anyone have any experience or knowledge regarding this?
GiteaMirror commented 2026-04-12 10:02:58 -05:00
Author
Owner
Copy Link

@wuyongyi commented on GitHub (Oct 18, 2023):

BTW: I use
litellm --model ollama/codellama
setup openai API compatible server for local LLMs in http://0.0.0.0:8000
as (https://docs.litellm.ai/docs/providers/ollama)
The test code for litellm works fine.
But I get error in open-interpreter:
"""

model, custom_llm_provider = get_llm_provider(model=model, custom_llm_provider=custom_llm_provider, api_base=api_base)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\anaconda3\Lib\site-packages\litellm\utils.py", line 1483, in get_llm_provider
raise e
File "D:\ProgramData\anaconda3\Lib\site-packages\litellm\utils.py", line 1480, in get_llm_provider
raise ValueError(f"LLM Provider NOT provided. Pass in the LLM provider you are trying to call. E.g. For 'Huggingface' inference endpoints pass in completion(model='huggingface/{model}',..) Learn more: https://docs.litellm.ai/docs/providers")

...

raise APIError(status_code=500, message=str(original_exception), llm_provider=custom_llm_provider, model=model)
litellm.exceptions.APIError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. E.g. For 'Huggingface' inference endpoints pass in completion(model='huggingface/gpt-4',..) Learn more: https://docs.litellm.ai/docs/providers
"""

<!-- gh-comment-id:1768458083 --> @wuyongyi commented on GitHub (Oct 18, 2023): BTW: I use litellm --model ollama/codellama setup openai API compatible server for local LLMs in http://0.0.0.0:8000 as (https://docs.litellm.ai/docs/providers/ollama) The test code for litellm works fine. But I get error in open-interpreter: """ model, custom_llm_provider = get_llm_provider(model=model, custom_llm_provider=custom_llm_provider, api_base=api_base) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ProgramData\anaconda3\Lib\site-packages\litellm\utils.py", line 1483, in get_llm_provider raise e File "D:\ProgramData\anaconda3\Lib\site-packages\litellm\utils.py", line 1480, in get_llm_provider raise ValueError(f"LLM Provider NOT provided. Pass in the LLM provider you are trying to call. E.g. For 'Huggingface' inference endpoints pass in `completion(model='huggingface/{model}',..)` Learn more: https://docs.litellm.ai/docs/providers") ... raise APIError(status_code=500, message=str(original_exception), llm_provider=custom_llm_provider, model=model) litellm.exceptions.APIError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. E.g. For 'Huggingface' inference endpoints pass in `completion(model='huggingface/gpt-4',..)` Learn more: https://docs.litellm.ai/docs/providers """
GiteaMirror commented 2026-04-12 10:02:58 -05:00
Author
Owner
Copy Link

@wuyongyi commented on GitHub (Oct 18, 2023):

update:
run :
interpreter --model openai/codellama --api_base http://127.0.0.1:8000/
now the connection is fine.

<!-- gh-comment-id:1769530696 --> @wuyongyi commented on GitHub (Oct 18, 2023): update: run : interpreter --model openai/codellama --api_base http://127.0.0.1:8000/ now the connection is fine.
GiteaMirror commented 2026-04-12 10:02:58 -05:00
Author
Owner
Copy Link

@ghost commented on GitHub (Oct 20, 2023):

hey @wuyongyi openinterpreter uses litellm, so you could just use it as ollama/codellama in the interpreter call

<!-- gh-comment-id:1773436863 --> @ghost commented on GitHub (Oct 20, 2023): hey @wuyongyi openinterpreter uses litellm, so you could just use it as ollama/codellama in the interpreter call
GiteaMirror commented 2026-04-12 10:02:59 -05:00
Author
Owner
Copy Link

@Axenide commented on GitHub (Nov 17, 2023):

hey @wuyongyi openinterpreter uses litellm, so you could just use it as ollama/codellama in the interpreter call

How did you do it? I tried with interpreter --model ollama/codellama --api_base http://127.0.0.1:11434/ but I get this error:

▌ Model set to ollama/codellama                                                                                                                                                                                                                                          

Open Interpreter will require approval before running code.                                                                                                                                                                                                                

Use interpreter -y to bypass this.                                                                                                                                                                                                                                         

Press CTRL-C to exit.                                                                                                                                                                                                                                                      

> Hello there!

We were unable to determine the context window of this model. Defaulting to 3000.                                                                                                                                                                                          
If your model can handle more, run interpreter --context_window {token limit} or interpreter.context_window = {token limit}.                                                                                                                                               
Also, please set max_tokens: interpreter --max_tokens {max tokens per response} or interpreter.max_tokens = {max tokens per response}                                                                                                                                      

Traceback (most recent call last):
  File "/home/adriano/.local/pipx/venvs/open-interpreter/lib/python3.11/site-packages/litellm/llms/ollama.py", line 133, in get_ollama_response_stream
    j = json.loads(chunk)
        ^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/json/__init__.py", line 346, in loads
    return _default_decoder.decode(s)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/json/decoder.py", line 340, in decode
    raise JSONDecodeError("Extra data", s, end)
json.decoder.JSONDecodeError: Extra data: line 1 column 5 (char 4)
<!-- gh-comment-id:1815711609 --> @Axenide commented on GitHub (Nov 17, 2023): > hey @wuyongyi openinterpreter uses litellm, so you could just use it as ollama/codellama in the interpreter call How did you do it? I tried with `interpreter --model ollama/codellama --api_base http://127.0.0.1:11434/` but I get this error: ``` ▌ Model set to ollama/codellama Open Interpreter will require approval before running code. Use interpreter -y to bypass this. Press CTRL-C to exit. > Hello there! We were unable to determine the context window of this model. Defaulting to 3000. If your model can handle more, run interpreter --context_window {token limit} or interpreter.context_window = {token limit}. Also, please set max_tokens: interpreter --max_tokens {max tokens per response} or interpreter.max_tokens = {max tokens per response} Traceback (most recent call last): File "/home/adriano/.local/pipx/venvs/open-interpreter/lib/python3.11/site-packages/litellm/llms/ollama.py", line 133, in get_ollama_response_stream j = json.loads(chunk) ^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.11/json/__init__.py", line 346, in loads return _default_decoder.decode(s) ^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.11/json/decoder.py", line 340, in decode raise JSONDecodeError("Extra data", s, end) json.decoder.JSONDecodeError: Extra data: line 1 column 5 (char 4) ```
GiteaMirror commented 2026-04-12 10:02:59 -05:00
Author
Owner
Copy Link

@tokutake commented on GitHub (Nov 22, 2023):

@Axenide
I encountered the same error you did.

[GIN] 2023/11/22 - 21:03:44 | 404 |       2.256µs |       127.0.0.1 | POST     "//api/generate"

There was an extra slash in the request URL.

I removed the trailing slash from the api_base argument URL and the error went away.

> interpreter --model ollama/llama2 --api_base http://localhost:11434

> hi

We were unable to determine the context window of this model. Defaulting to 3000.                                             
If your model can handle more, run interpreter --context_window {token limit} or interpreter.context_window = {token limit}.  
Also, please set max_tokens: interpreter --max_tokens {max tokens per response} or interpreter.max_tokens = {max tokens per   
response}                                                                                                                     

b'{"model":"llama2","created_at":"2023-11-22T12:11:27.893947Z","response":"Hello","done":false}'
<!-- gh-comment-id:1822658989 --> @tokutake commented on GitHub (Nov 22, 2023): @Axenide I encountered the same error you did. ``` [GIN] 2023/11/22 - 21:03:44 | 404 | 2.256µs | 127.0.0.1 | POST "//api/generate" ``` There was an extra slash in the request URL. I removed the trailing slash from the api_base argument URL and the error went away. ``` > interpreter --model ollama/llama2 --api_base http://localhost:11434 > hi We were unable to determine the context window of this model. Defaulting to 3000. If your model can handle more, run interpreter --context_window {token limit} or interpreter.context_window = {token limit}. Also, please set max_tokens: interpreter --max_tokens {max tokens per response} or interpreter.max_tokens = {max tokens per response} b'{"model":"llama2","created_at":"2023-11-22T12:11:27.893947Z","response":"Hello","done":false}' ```
GiteaMirror commented 2026-04-12 10:03:00 -05:00
Author
Owner
Copy Link

@Axenide commented on GitHub (Nov 22, 2023):

@Axenide
I encountered the same error you did.

[GIN] 2023/11/22 - 21:03:44 | 404 |       2.256µs |       127.0.0.1 | POST     "//api/generate"

There was an extra slash in the request URL.

I removed the trailing slash from the api_base argument URL and the error went away.

> interpreter --model ollama/llama2 --api_base http://localhost:11434

> hi

We were unable to determine the context window of this model. Defaulting to 3000.                                             
If your model can handle more, run interpreter --context_window {token limit} or interpreter.context_window = {token limit}.  
Also, please set max_tokens: interpreter --max_tokens {max tokens per response} or interpreter.max_tokens = {max tokens per   
response}                                                                                                                     

b'{"model":"llama2","created_at":"2023-11-22T12:11:27.893947Z","response":"Hello","done":false}'

Actually, the solution was to specify the version of the model. :)

ollama/codellama:latest

<!-- gh-comment-id:1822799610 --> @Axenide commented on GitHub (Nov 22, 2023): > @Axenide > I encountered the same error you did. > > ``` > [GIN] 2023/11/22 - 21:03:44 | 404 | 2.256µs | 127.0.0.1 | POST "//api/generate" > ``` > There was an extra slash in the request URL. > > I removed the trailing slash from the api_base argument URL and the error went away. > > ``` > > interpreter --model ollama/llama2 --api_base http://localhost:11434 > > > hi > > We were unable to determine the context window of this model. Defaulting to 3000. > If your model can handle more, run interpreter --context_window {token limit} or interpreter.context_window = {token limit}. > Also, please set max_tokens: interpreter --max_tokens {max tokens per response} or interpreter.max_tokens = {max tokens per > response} > > b'{"model":"llama2","created_at":"2023-11-22T12:11:27.893947Z","response":"Hello","done":false}' > ``` > > Actually, the solution was to specify the version of the model. :) `ollama/codellama:latest`
GiteaMirror commented 2026-04-12 10:03:00 -05:00
Author
Owner
Copy Link

@phrane commented on GitHub (Dec 31, 2023):

@Axenide
I'm still getting the same errors previously described despite specifying the model version (as displayed in 'Ollama list' cmd). I'm on intel-mac, running mistral, and here is what I used:
interpreter --model ollama/mistral:latest --api_base http://127.0.0.1:11434/
but I then get the following from open interpreter:

▌ Model set to openai/ollama/mistral:latest

Open Interpreter will require approval before running code.
Use interpreter -y to bypass this.
Press CTRL-C to exit.

...i.e. openai prefixed on the model selection. Any ideas on how to correct this?

<!-- gh-comment-id:1872983185 --> @phrane commented on GitHub (Dec 31, 2023): @Axenide I'm still getting the same errors previously described despite specifying the model version (as displayed in 'Ollama list' cmd). I'm on intel-mac, running mistral, and here is what I used: `interpreter --model ollama/mistral:latest --api_base http://127.0.0.1:11434/` but I then get the following from open interpreter: ``` ▌ Model set to openai/ollama/mistral:latest Open Interpreter will require approval before running code. Use interpreter -y to bypass this. Press CTRL-C to exit. ``` ...i.e. openai prefixed on the model selection. Any ideas on how to correct this?
GiteaMirror commented 2026-04-12 10:03:00 -05:00
Author
Owner
Copy Link

@Axenide commented on GitHub (Dec 31, 2023):

@phrane
Try without ollama/, it used to work when specifying it as openai, so I guess it should work if you just use mistral:latest.

<!-- gh-comment-id:1873001362 --> @Axenide commented on GitHub (Dec 31, 2023): @phrane Try without `ollama/`, it used to work when specifying it as openai, so I guess it should work if you just use `mistral:latest`.
GiteaMirror commented 2026-04-12 10:03:01 -05:00
Author
Owner
Copy Link

@ghost commented on GitHub (Jan 1, 2024):

@phrane that looks wrong. I think openinterpreter is automatically assuming this is an openai compatible endpoint (see the appended openai/),

LiteLLM (which openinterpreter uses to make api calls) supports ollama/mistral and that should work fine - https://docs.litellm.ai/docs/providers/ollama.

But it seems like open-interpreter hasn't added support yet - https://docs.openinterpreter.com/language-model-setup/hosted-models/openai

Would recommend filing an issue on their github - https://github.com/KillianLucas/open-interpreter

<!-- gh-comment-id:1873089125 --> @ghost commented on GitHub (Jan 1, 2024): @phrane that looks wrong. I think openinterpreter is automatically assuming this is an openai compatible endpoint (see the appended `openai/`), LiteLLM (which openinterpreter uses to make api calls) supports `ollama/mistral` and that should work fine - https://docs.litellm.ai/docs/providers/ollama. But it seems like open-interpreter hasn't added support yet - https://docs.openinterpreter.com/language-model-setup/hosted-models/openai Would recommend filing an issue on their github - https://github.com/KillianLucas/open-interpreter
GiteaMirror commented 2026-04-12 10:03:01 -05:00
Author
Owner
Copy Link

@hidek84 commented on GitHub (Jan 1, 2024):

You no longer need the --api-base argument when connecting to a local Ollama model endpoint because LiteLLM will automatically handle it.

So, I think you can try the following command (open-interpreter==0.1.18 and litellm==1.16.7):

interpreter --model ollama/mistral:latest

For more details, please refer to my comment in the open-interpreter's GitHub issue below.
https://github.com/KillianLucas/open-interpreter/issues/856#issuecomment-1872673035

<!-- gh-comment-id:1873363488 --> @hidek84 commented on GitHub (Jan 1, 2024): You no longer need the `--api-base` argument when connecting to a local Ollama model endpoint because LiteLLM will automatically handle it. So, I think you can try the following command (`open-interpreter==0.1.18` and `litellm==1.16.7`): ``` interpreter --model ollama/mistral:latest ``` For more details, please refer to my comment in the open-interpreter's GitHub issue below. https://github.com/KillianLucas/open-interpreter/issues/856#issuecomment-1872673035
GiteaMirror commented 2026-04-12 10:03:02 -05:00
Author
Owner
Copy Link

@phrane commented on GitHub (Jan 1, 2024):

@hidek84 this works a treat, thanks!!
I still get a msg:

If your model can handle more, run `interpreter --context_window {token limit}` or `interpreter.llm.context_window = {token limit}`.
Also, please set max_tokens: `interpreter --max_tokens {max tokens per response}` or `interpreter.llm.max_tokens = {max tokens per response}`

but I'm guessing this is due to a config I need to update somewhere.

<!-- gh-comment-id:1873373092 --> @phrane commented on GitHub (Jan 1, 2024): @hidek84 this works a treat, thanks!! I still get a msg: ```We were unable to determine the context window of this model. Defaulting to 3000. If your model can handle more, run `interpreter --context_window {token limit}` or `interpreter.llm.context_window = {token limit}`. Also, please set max_tokens: `interpreter --max_tokens {max tokens per response}` or `interpreter.llm.max_tokens = {max tokens per response}` ``` but I'm guessing this is due to a config I need to update somewhere.
GiteaMirror commented 2026-04-12 10:03:02 -05:00
Author
Owner
Copy Link

@phrane commented on GitHub (Jan 1, 2024):

@phrane that looks wrong. I think openinterpreter is automatically assuming this is an openai compatible endpoint (see the appended openai/),

LiteLLM (which openinterpreter uses to make api calls) supports ollama/mistral and that should work fine - https://docs.litellm.ai/docs/providers/ollama.

But it seems like open-interpreter hasn't added support yet - https://docs.openinterpreter.com/language-model-setup/hosted-models/openai

Would recommend filing an issue on their github - https://github.com/KillianLucas/open-interpreter

@krrishdholakia thanks for your input bro, my thinking was in line with yours regarding open default assumption. Was about to log an issue, but for @hidek84 . Thanks alll round!!

<!-- gh-comment-id:1873374262 --> @phrane commented on GitHub (Jan 1, 2024): > @phrane that looks wrong. I think openinterpreter is automatically assuming this is an openai compatible endpoint (see the appended `openai/`), > > LiteLLM (which openinterpreter uses to make api calls) supports `ollama/mistral` and that should work fine - https://docs.litellm.ai/docs/providers/ollama. > > But it seems like open-interpreter hasn't added support yet - https://docs.openinterpreter.com/language-model-setup/hosted-models/openai > > Would recommend filing an issue on their github - https://github.com/KillianLucas/open-interpreter @krrishdholakia thanks for your input bro, my thinking was in line with yours regarding open default assumption. Was about to log an issue, but for @hidek84 . Thanks alll round!!
GiteaMirror commented 2026-04-12 10:03:03 -05:00
Author
Owner
Copy Link

@obadiahpewee commented on GitHub (Mar 14, 2025):

i'm still get issues with this, trying to connect via ollama on debian/linux
interpreter --model ollama/qwen2.5:7b-instruct-q8_0

▌ Model set to ollama/qwen2.5:7b-instruct-q8_0

Loading qwen2.5:7b-instruct-q8_0...

Model loaded.

Open Interpreter will require approval before running code.

Use interpreter -y to bypass this.

Press CTRL-C to exit.

hello

{"name": "execute", "arguments": {"language": "python", "code": "print('Hello, obione!')"}}

how are you

{"name": "execute", "arguments": {"language": "python", "code": "print(f'Hello, {obione}! How can I assist you today?')"}}

what os is this on

{"name": "execute", "arguments": {"language": "shell", "code": "uname"}}

what is the time in LA,USA right now

{"name": "execute", "arguments": {"language": "python", "code": "import platform\nprint(platform.system())"}}

Exiting...

<!-- gh-comment-id:2723851709 --> @obadiahpewee commented on GitHub (Mar 14, 2025): i'm still get issues with this, trying to connect via ollama on debian/linux interpreter --model ollama/qwen2.5:7b-instruct-q8_0 ▌ Model set to ollama/qwen2.5:7b-instruct-q8_0 Loading qwen2.5:7b-instruct-q8_0... Model loaded. Open Interpreter will require approval before running code. Use interpreter -y to bypass this. Press CTRL-C to exit. > hello {"name": "execute", "arguments": {"language": "python", "code": "print('Hello, obione!')"}} > how are you {"name": "execute", "arguments": {"language": "python", "code": "print(f'Hello, {obione}! How can I assist you today?')"}} > what os is this on {"name": "execute", "arguments": {"language": "shell", "code": "uname"}} > what is the time in LA,USA right now {"name": "execute", "arguments": {"language": "python", "code": "import platform\nprint(platform.system())"}} > Exiting...
GiteaMirror commented 2026-04-12 10:03:04 -05:00
Author
Owner
Copy Link

@obadiahpewee commented on GitHub (Mar 14, 2025):

same result with --local flag used
interpreter --local

Open Interpreter supports multiple local model providers.

[?] Select a provider:

Ollama
Llamafile
LM Studio
Jan

[?] Select a model:
llama3.2

qwen2.5:7b-instruct-q8_0
qwen2.5:32b
qwen2.5-coder:32b
huihui_ai/qwen2.5-1m-abliterated:14b-instruct-q8_0
deepseek-r1:32b
↓ Download llama3.1
↓ Download phi3
↓ Download mistral-nemo
↓ Download gemma2
↓ Download codestral
Browse Models ↗

Loading qwen2.5:7b-instruct-q8_0...

Model loaded.

▌ Model set to qwen2.5:7b-instruct-q8_0

Open Interpreter will require approval before running code.

Use interpreter -y to bypass this.

Press CTRL-C to exit.

hello

{"name": "execute", "arguments": {"language": "python", "code": "print('hello')"}}

how are you

{"name": "execute", "arguments": {"language": "python", "code": "print('How are you?')"}}

time in UK

{"name": "execute", "arguments": {"language": "python", "code": "from datetime import datetime; print(datetime.now().strftime('%Y-%m-%d
%H:%M:%S %Z'))"}}

answer?

{"name": "execute", "arguments": {"language": "python", "code": "print('The current time in UK is',
datetime.now(pytz.timezone('Europe/London')).strftime('%Y-%m-%d %H:%M:%S'))"}}

<!-- gh-comment-id:2723862651 --> @obadiahpewee commented on GitHub (Mar 14, 2025): same result with --local flag used interpreter --local Open Interpreter supports multiple local model providers. [?] Select a provider: > Ollama Llamafile LM Studio Jan [?] Select a model: llama3.2 > qwen2.5:7b-instruct-q8_0 qwen2.5:32b qwen2.5-coder:32b huihui_ai/qwen2.5-1m-abliterated:14b-instruct-q8_0 deepseek-r1:32b ↓ Download llama3.1 ↓ Download phi3 ↓ Download mistral-nemo ↓ Download gemma2 ↓ Download codestral Browse Models ↗ Loading qwen2.5:7b-instruct-q8_0... Model loaded. ▌ Model set to qwen2.5:7b-instruct-q8_0 Open Interpreter will require approval before running code. Use interpreter -y to bypass this. Press CTRL-C to exit. > hello {"name": "execute", "arguments": {"language": "python", "code": "print('hello')"}} > how are you {"name": "execute", "arguments": {"language": "python", "code": "print('How are you?')"}} > time in UK {"name": "execute", "arguments": {"language": "python", "code": "from datetime import datetime; print(datetime.now().strftime('%Y-%m-%d %H:%M:%S %Z'))"}} > answer? {"name": "execute", "arguments": {"language": "python", "code": "print('The current time in UK is', datetime.now(pytz.timezone('Europe/London')).strftime('%Y-%m-%d %H:%M:%S'))"}}
GiteaMirror commented 2026-04-12 10:03:04 -05:00
Author
Owner
Copy Link

@trevorstr commented on GitHub (Mar 15, 2025):

I'm trying to run the development branch with this:

interpreter --api-base http://myserver.local:11434 --model llama3.2-vision:11b --provider ollama

And then when I enter a prompt, I get:

Open Interpreter 1.0.0
Copyright (C) 2024 Open Interpreter Team
Licensed under GNU AGPL v3.0

A modern command-line assistant.

Usage: i [prompt]
   or: interpreter [options]

Documentation: docs.openinterpreter.com
Run 'interpreter --help' for all options

> asdf

   Traceback (most recent call last):
  File "/Users/trevor.sullivan/.local/bin/interpreter", line 8, in <module>
    sys.exit(main())
             ~~~~^^
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/interpreter/cli.py", line 306, in main
    asyncio.run(async_main(args))
    ~~~~~~~~~~~^^^^^^^^^^^^^^^^^^
  File "/opt/homebrew/Cellar/python@3.13/3.13.2/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/runners.py", line 195, in run
    return runner.run(main)
           ~~~~~~~~~~^^^^^^
  File "/opt/homebrew/Cellar/python@3.13/3.13.2/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^
  File "/opt/homebrew/Cellar/python@3.13/3.13.2/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/base_events.py", line 725, in run_until_complete
    return future.result()
           ~~~~~~~~~~~~~^^
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/interpreter/cli.py", line 230, in async_main
    async for _ in global_interpreter.async_respond():
        pass
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/interpreter/interpreter.py", line 740, in async_respond
    raw_response = litellm.completion(**params)
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/utils.py", line 1235, in wrapper
    raise e
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/utils.py", line 1113, in wrapper
    result = original_function(*args, **kwargs)
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/main.py", line 3101, in completion
    raise exception_type(
    ...<5 lines>...
    )
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/main.py", line 984, in completion
    model, custom_llm_provider, dynamic_api_key, api_base = get_llm_provider(
                                                            ~~~~~~~~~~~~~~~~^
        model=model,
        ^^^^^^^^^^^^
    ...<2 lines>...
        api_key=api_key,
        ^^^^^^^^^^^^^^^^
    )
    ^
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/litellm_core_utils/get_llm_provider_logic.py", line 356, in get_llm_provider
    raise e
  File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/litellm_core_utils/get_llm_provider_logic.py", line 333, in get_llm_provider
    raise litellm.exceptions.BadRequestError(  # type: ignore
    ...<8 lines>...
    )
litellm.exceptions.BadRequestError: litellm.BadRequestError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. You passed model=llama3.2-vision:11b
 Pass model as E.g. For 'Huggingface' inference endpoints pass in `completion(model='huggingface/starcoder',..)` Learn more: https://docs.litellm.ai/docs/providers
<!-- gh-comment-id:2726089256 --> @trevorstr commented on GitHub (Mar 15, 2025): I'm trying to run the development branch with this: ``` interpreter --api-base http://myserver.local:11434 --model llama3.2-vision:11b --provider ollama ``` And then when I enter a prompt, I get: ``` Open Interpreter 1.0.0 Copyright (C) 2024 Open Interpreter Team Licensed under GNU AGPL v3.0 A modern command-line assistant. Usage: i [prompt] or: interpreter [options] Documentation: docs.openinterpreter.com Run 'interpreter --help' for all options > asdf Traceback (most recent call last): File "/Users/trevor.sullivan/.local/bin/interpreter", line 8, in <module> sys.exit(main()) ~~~~^^ File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/interpreter/cli.py", line 306, in main asyncio.run(async_main(args)) ~~~~~~~~~~~^^^^^^^^^^^^^^^^^^ File "/opt/homebrew/Cellar/python@3.13/3.13.2/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/runners.py", line 195, in run return runner.run(main) ~~~~~~~~~~^^^^^^ File "/opt/homebrew/Cellar/python@3.13/3.13.2/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/runners.py", line 118, in run return self._loop.run_until_complete(task) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^ File "/opt/homebrew/Cellar/python@3.13/3.13.2/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/base_events.py", line 725, in run_until_complete return future.result() ~~~~~~~~~~~~~^^ File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/interpreter/cli.py", line 230, in async_main async for _ in global_interpreter.async_respond(): pass File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/interpreter/interpreter.py", line 740, in async_respond raw_response = litellm.completion(**params) File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/utils.py", line 1235, in wrapper raise e File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/utils.py", line 1113, in wrapper result = original_function(*args, **kwargs) File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/main.py", line 3101, in completion raise exception_type( ...<5 lines>... ) File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/main.py", line 984, in completion model, custom_llm_provider, dynamic_api_key, api_base = get_llm_provider( ~~~~~~~~~~~~~~~~^ model=model, ^^^^^^^^^^^^ ...<2 lines>... api_key=api_key, ^^^^^^^^^^^^^^^^ ) ^ File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/litellm_core_utils/get_llm_provider_logic.py", line 356, in get_llm_provider raise e File "/Users/trevor.sullivan/.local/pipx/venvs/open-interpreter/lib/python3.13/site-packages/litellm/litellm_core_utils/get_llm_provider_logic.py", line 333, in get_llm_provider raise litellm.exceptions.BadRequestError( # type: ignore ...<8 lines>... ) litellm.exceptions.BadRequestError: litellm.BadRequestError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. You passed model=llama3.2-vision:11b Pass model as E.g. For 'Huggingface' inference endpoints pass in `completion(model='huggingface/starcoder',..)` Learn more: https://docs.litellm.ai/docs/providers ```
GiteaMirror commented 2026-04-12 10:03:05 -05:00
Author
Owner
Copy Link

@ghost commented on GitHub (Mar 15, 2025):

hey @trevorstr looks like the provider isn't being passed in

try --model ollama/llama3.2-vision:11b

<!-- gh-comment-id:2726183460 --> @ghost commented on GitHub (Mar 15, 2025): hey @trevorstr looks like the provider isn't being passed in try `--model ollama/llama3.2-vision:11b`
GiteaMirror commented 2026-04-12 10:03:06 -05:00
Author
Owner
Copy Link

@trevorstr commented on GitHub (Mar 15, 2025):

@krrishdholakia I tried that first, and it still did not work.

I was using the development branch. Are you using that branch as well?

<!-- gh-comment-id:2726891295 --> @trevorstr commented on GitHub (Mar 15, 2025): @krrishdholakia I tried that first, and it still did not work. I was using the `development` branch. Are you using that branch as well?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#402
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?