Fix: remove 4o reasoning labels; gate reasoning_effort to supported models #34
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…dels
The site was generating pages like “ChatGPT-4o (Medium/High Reasoning)”
and passing a reasoning_effort parameter to 4o. GPT‑4o is not a
reasoning model, so these labels were misleading and suggested eval
claims that aren’t true. Passing reasoning_effort to unsupported models
is also incorrect and can lead to API errors or undefined behavior.
Changes
- assess.py:
- Remove “ChatGPT-4o (Medium/High Reasoning)” entries; expose a single
canonical “ChatGPT-4o”.
- Drop legacy normalization hook to avoid silently rewriting model keys.
- models/openai.py:
- Add ALLOWED_REASONING_MODELS and only include reasoning_effort for
models that support it (O-series + GPT‑5 family).
- Include GPT‑5 variants in the allowed list.
- Refactor request construction to conditionally attach
reasoning_effort via kwargs.
- Minor formatting of SKIP_TEMPERATURE for readability.
Impact
- Eliminates misleading 4o “reasoning” pages/URLs and labels.
- Prevents sending reasoning_effort to unsupported models.
- Keeps 4o results page without any reasoning level.
- No change to evaluation logic; only labeling/parameter hygiene.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR removes misleading GPT‑4o “reasoning” labels and ensures
reasoning_effortis only sent to models that support it.Why
reasoning_effortto unsupported models is incorrect and may cause API errors.What
ALLOWED_REASONING_MODELS(O‑series + GPT‑5) and gatereasoning_effort; minor refactor/formatting.Impact
reasoning_effortsent to unsupported models.