OpenHands icon indicating copy to clipboard operation
OpenHands copied to clipboard
verified publisher icon All-Hands-AI

[Bug]: 'temperature' does not support 0.0 with this model (O1-mini)

Open AlexCuadron opened this issue 1 year ago • 11 comments

Is there an existing issue for the same bug?

  • [X] I have checked the troubleshooting document at https://docs.all-hands.dev/modules/usage/troubleshooting
  • [X] I have checked the existing issues.

Describe the bug

I tried running OpenHands with O1-mini as my default model and the model keeps "encountering an error" as described by the UI. However, upon closer inspection of the terminal. This seems to be the problem: litellm.exceptions.BadRequestError: litellm.BadRequestError: OpenAIException - Error code: 400 - {'error': {'message': "Unsupported value: 'temperature' does not support 0.0 with this model. Only the default (1) value is supported.", 'type': 'invalid_request_error', 'param': 'temperature', 'code': 'unsupported_value'}}

Current OpenHands version

I am on the main branch, at commit: b4626ab93ebac96d87d239432f49f8af44cd0744

Installation and Configuration

git clone https://github.com/AlexCuadron/OpenHands.git
cd OpenHands
git remote add upstream [email protected]:All-Hands-AI/OpenHands.git
git remote -v
git fetch upstream\ngit checkout main\ngit merge upstream/main\ngit push origin main
conda create --name open-hands python=3.11 conda-forge::nodejs conda-forge::poetry
conda activate open-hands
brew install netcat
pip install git+https://github.com/OpenDevin/SWE-bench.git@7b0c4b1c249ed4b4600a5bba8afb916d543e034a
make build
poetry self update
make build
make setup-config
Setting up config.toml...
Enter your workspace directory (as absolute path) [default: ./workspace]:
Enter your LLM model name, used for running without UI. Set the model in the UI after you start the app. (see https://docs.litellm.ai/docs/providers for full list) [default: gpt-4o]: o1-mini
Enter your LLM api key:
Enter your LLM base URL [mostly used for local LLMs, leave blank if not needed - example: http://localhost:5001/v1/]:
Enter your LLM Embedding Model
Choices are:
  - openai
  - azureopenai
  - Embeddings available only with OllamaEmbedding:
    - llama2
    - mxbai-embed-large
    - nomic-embed-text
    - all-minilm
    - stable-code
    - bge-m3
    - bge-large
    - paraphrase-multilingual
    - snowflake-arctic-embed
  - Leave blank to default to 'BAAI/bge-small-en-v1.5' via huggingface
>
Config.toml setup completed.

export DEBUG=1

Model and Agent

  • o1-mini

Operating System

MacOS

Reproduction Steps

  1. Open OpenHands
  2. Ask the LLM what model are you?
  3. The agent will find an error.

Logs, Errors, Screenshots, and Additional Context

==============
CodeActAgent LEVEL 0 LOCAL STEP 0 GLOBAL STEP 0

17:36:41 - openhands:DEBUG: logger.py:238 - Logging to /Users/acuadron/Documents/OpenDevin/OpenHands/logs/llm/24-09-30_17-30/prompt_002.log

Give Feedback / Get Help: https://github.com/BerriAI/litellm/issues/new
LiteLLM.Info: If you need to debug this error, use `litellm.set_verbose=True'.

Traceback (most recent call last):
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 778, in completion
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 714, in completion
    self.make_sync_openai_chat_completion_request(
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 573, in make_sync_openai_chat_completion_request
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 562, in make_sync_openai_chat_completion_request
    raw_response = openai_client.chat.completions.with_raw_response.create(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_legacy_response.py", line 353, in wrapped
    return cast(LegacyAPIResponse[R], func(*args, **kwargs))
                                      ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_utils/_utils.py", line 274, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/resources/chat/completions.py", line 704, in create
    return self._post(
           ^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_base_client.py", line 1270, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_base_client.py", line 947, in request
    return self._request(
           ^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_base_client.py", line 1051, in _request
    raise self._make_status_error_from_response(err.response) from None
openai.BadRequestError: Error code: 400 - {'error': {'message': "Unsupported value: 'temperature' does not support 0.0 with this model. Only the default (1) value is supported.", 'type': 'invalid_request_error', 'param': 'temperature', 'code': 'unsupported_value'}}

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/main.py", line 1470, in completion
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/main.py", line 1423, in completion
    response = openai_o1_chat_completions.completion(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/chat/o1_handler.py", line 58, in completion
    response = super().completion(
               ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 788, in completion
    raise OpenAIError(
litellm.llms.OpenAI.openai.OpenAIError: Error code: 400 - {'error': {'message': "Unsupported value: 'temperature' does not support 0.0 with this model. Only the default (1) value is supported.", 'type': 'invalid_request_error', 'param': 'temperature', 'code': 'unsupported_value'}}

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/acuadron/Documents/OpenDevin/OpenHands/openhands/controller/agent_controller.py", line 155, in start_step_loop
    await self._step()
  File "/Users/acuadron/Documents/OpenDevin/OpenHands/openhands/controller/agent_controller.py", line 422, in _step
    action = self.agent.step(self.state)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Documents/OpenDevin/OpenHands/agenthub/codeact_agent/codeact_agent.py", line 207, in step
    response = self.llm.completion(**params)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 336, in wrapped_f
    return copy(f, *args, **kw)
           ^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 475, in __call__
    do = self.iter(retry_state=retry_state)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 376, in iter
    result = action(retry_state)
             ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 398, in <lambda>
    self._add_action_func(lambda rs: rs.outcome.result())
                                     ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/miniconda3/envs/open-hands/lib/python3.11/concurrent/futures/_base.py", line 449, in result
    return self.__get_result()
           ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/miniconda3/envs/open-hands/lib/python3.11/concurrent/futures/_base.py", line 401, in __get_result
    raise self._exception
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 478, in __call__
    result = fn(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Documents/OpenDevin/OpenHands/openhands/llm/llm.py", line 187, in wrapper
    resp: ModelResponse = completion_unwrapped(*args, **kwargs)
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/utils.py", line 1058, in wrapper
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/utils.py", line 946, in wrapper
    result = original_function(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/main.py", line 2900, in completion
    raise exception_type(
          ^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/litellm_core_utils/exception_mapping_utils.py", line 2033, in exception_type
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/litellm_core_utils/exception_mapping_utils.py", line 297, in exception_type
    raise BadRequestError(
litellm.exceptions.BadRequestError: litellm.BadRequestError: OpenAIException - Error code: 400 - {'error': {'message': "Unsupported value: 'temperature' does not support 0.0 with this model. Only the default (1) value is supported.", 'type': 'invalid_request_error', 'param': 'temperature', 'code': 'unsupported_value'}}
17:36:42 - openhands:ERROR: agent_controller.py:161 - Error while running the agent: litellm.BadRequestError: OpenAIException - Error code: 400 - {'error': {'message': "Unsupported value: 'temperature' does not support 0.0 with this model. Only the default (1) value is supported.", 'type': 'invalid_request_error', 'param': 'temperature', 'code': 'unsupported_value'}}
17:36:42 - openhands:ERROR: agent_controller.py:162 - Traceback (most recent call last):
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 778, in completion
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 714, in completion
    self.make_sync_openai_chat_completion_request(
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 573, in make_sync_openai_chat_completion_request
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 562, in make_sync_openai_chat_completion_request
    raw_response = openai_client.chat.completions.with_raw_response.create(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_legacy_response.py", line 353, in wrapped
    return cast(LegacyAPIResponse[R], func(*args, **kwargs))
                                      ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_utils/_utils.py", line 274, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/resources/chat/completions.py", line 704, in create
    return self._post(
           ^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_base_client.py", line 1270, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_base_client.py", line 947, in request
    return self._request(
           ^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/openai/_base_client.py", line 1051, in _request
    raise self._make_status_error_from_response(err.response) from None
openai.BadRequestError: Error code: 400 - {'error': {'message': "Unsupported value: 'temperature' does not support 0.0 with this model. Only the default (1) value is supported.", 'type': 'invalid_request_error', 'param': 'temperature', 'code': 'unsupported_value'}}

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/main.py", line 1470, in completion
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/main.py", line 1423, in completion
    response = openai_o1_chat_completions.completion(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/chat/o1_handler.py", line 58, in completion
    response = super().completion(
               ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/llms/OpenAI/openai.py", line 788, in completion
    raise OpenAIError(
litellm.llms.OpenAI.openai.OpenAIError: Error code: 400 - {'error': {'message': "Unsupported value: 'temperature' does not support 0.0 with this model. Only the default (1) value is supported.", 'type': 'invalid_request_error', 'param': 'temperature', 'code': 'unsupported_value'}}

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/acuadron/Documents/OpenDevin/OpenHands/openhands/controller/agent_controller.py", line 155, in start_step_loop
    await self._step()
  File "/Users/acuadron/Documents/OpenDevin/OpenHands/openhands/controller/agent_controller.py", line 422, in _step
    action = self.agent.step(self.state)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Documents/OpenDevin/OpenHands/agenthub/codeact_agent/codeact_agent.py", line 207, in step
    response = self.llm.completion(**params)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 336, in wrapped_f
    return copy(f, *args, **kw)
           ^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 475, in __call__
    do = self.iter(retry_state=retry_state)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 376, in iter
    result = action(retry_state)
             ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 398, in <lambda>
    self._add_action_func(lambda rs: rs.outcome.result())
                                     ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/miniconda3/envs/open-hands/lib/python3.11/concurrent/futures/_base.py", line 449, in result
    return self.__get_result()
           ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/miniconda3/envs/open-hands/lib/python3.11/concurrent/futures/_base.py", line 401, in __get_result
    raise self._exception
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/tenacity/__init__.py", line 478, in __call__
    result = fn(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Documents/OpenDevin/OpenHands/openhands/llm/llm.py", line 187, in wrapper
    resp: ModelResponse = completion_unwrapped(*args, **kwargs)
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/utils.py", line 1058, in wrapper
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/utils.py", line 946, in wrapper
    result = original_function(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/main.py", line 2900, in completion
    raise exception_type(
          ^^^^^^^^^^^^^^^
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/litellm_core_utils/exception_mapping_utils.py", line 2033, in exception_type
    raise e
  File "/Users/acuadron/Library/Caches/pypoetry/virtualenvs/openhands-ai-RniBYVEh-py3.11/lib/python3.11/site-packages/litellm/litellm_core_utils/exception_mapping_utils.py", line 297, in exception_type
    raise BadRequestError(
litellm.exceptions.BadRequestError: litellm.BadRequestError: OpenAIException - Error code: 400 - {'error': {'message': "Unsupported value: 'temperature' does not support 0.0 with this model. Only the default (1) value is supported.", 'type': 'invalid_request_error', 'param': 'temperature', 'code': 'unsupported_value'}}

AlexCuadron avatar Oct 01 '24 00:10 AlexCuadron

@neubig I thought you had fixed something related to this?

mamoodi avatar Oct 01 '24 14:10 mamoodi

Hmm, I did think I fixed it in this: #4012

Let me see if I can repro.

neubig avatar Oct 01 '24 14:10 neubig

BTW @AlexCuadron , for the time being could you try

export LLM_TEMPERATURE=1

or add it to config.toml in the top OpenHands directory (you can see an example in config.template.toml).

neubig avatar Oct 01 '24 14:10 neubig

That works @neubig thx!

AlexCuadron avatar Oct 01 '24 18:10 AlexCuadron

Awesome! And let me re-open it, because we'd still like to fix this so it works without the workaround.

neubig avatar Oct 01 '24 18:10 neubig

I don't think there is something to fix in code and that the workaround is actually the solution. The #4012 PR takes care of unsupported params, but temperature is supported for e.g. other OpenAI models, just not o1. Thus, imo, setting the temperature on model-level is the correct way to go.

[llm.gpt-o1-mini]
temperature=1

tobitege avatar Oct 01 '24 21:10 tobitege

I see what you're saying, but that would mean that users who select o1-mini or o1-preview through the UI would have to first parse and understand the error message, find documentation about how to fix the problem (e.g. this issue), and then create a config.toml file with that setting. I think that UX-wise it'd be preferable for it to just work, so if they select o1-mini it will automatically default to temperature settings that work with o1 (possibly while throwing a warning to the user so they might notice that this happened).

neubig avatar Oct 02 '24 03:10 neubig

I agree that it ideally should work out of the box, I'm just wondering how to best deal with these "exceptions to the rule". Maybe we can have a separate, specific py file containing these special cases for arg/param changes, outside of the actual LLM class, so it's easy to find/maintain.

I'd assume that there are no plans to bring such LLM options into the UI in the near future or at all?

tobitege avatar Oct 02 '24 04:10 tobitege

I really like the idea of having all of these exceptions in a single place! I'm actually still a bit confused that litellm doesn't handle this though...

I think that having this sort of configuration in the UI might be worth it if we could think of a way to present it elegantly.

neubig avatar Oct 02 '24 04:10 neubig

I was working on this problem for a while and i got to know that the temperature setting is not available in the "o1-preview" or "o1-mini" models since it is still in the beta version. Like the previous comments, you cn set the temperature as 1 for now or you need not add the temperature parameter. It works either way. Plus it does not have tool compatability like(wikipideia or arxiv websraper).

I am attaching a link which has all the limitations of the model. Hopefully they add these functionalities in the future.

https://platform.openai.com/docs/guides/reasoning/quickstart

check this link and scroll down to beta limitations and see all its limitations

Aatif123-hub avatar Oct 18 '24 00:10 Aatif123-hub

This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.

github-actions[bot] avatar Nov 17 '24 02:11 github-actions[bot]

This issue was closed because it has been stalled for over 30 days with no activity.

github-actions[bot] avatar Nov 24 '24 02:11 github-actions[bot]

Any updates? How to set templerature=1? It is necessary for gpt-5 model

Webbrother avatar Aug 21 '25 10:08 Webbrother