fix: make model configurable via DASH_MODEL env var (default: gpt-4o)#17
Open
Arvuno wants to merge 1 commit into
Open
fix: make model configurable via DASH_MODEL env var (default: gpt-4o)#17Arvuno wants to merge 1 commit into
Arvuno wants to merge 1 commit into
Conversation
The hardcoded gpt-5.4 model ID doesn't exist and blocks all agent functionality. Extract to DASH_MODEL env var with safe default gpt-4o. Changes: - dash/settings.py: MODEL_ID from env var, MODEL uses MODEL_ID - evals/__init__.py: JUDGE_MODEL uses DASH_MODEL env var - evals/improve.py: get_improvement_plan uses DASH_MODEL env var - example.env: document DASH_MODEL - README.md: add DASH_MODEL to env vars table
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
The codebase had
gpt-5.4hardcoded as the model ID in three locations. GPT-5.4 does not exist in OpenAI's API, making Dash completely non-functional out of the box.This PR extracts the model ID into a
DASH_MODELenvironment variable with a safe default ofgpt-4o.Changes
dash/settings.pyMODEL_ID = getenv("DASH_MODEL", "gpt-4o")thenMODEL = OpenAIResponses(id=MODEL_ID)evals/__init__.pyJUDGE_MODELusesgetenv("DASH_MODEL", "gpt-4o")evals/improve.pyget_improvement_planusesgetenv("DASH_MODEL", "gpt-4o")for the bareOpenAI()client callexample.envDASH_MODEL=gpt-4oREADME.mdDASH_MODELto the Environment Variables tableTesting
N/A (no code logic changed, only config extraction).
Notes
evals/improve.pystill uses the bareOpenAI()client (as noted in Q-2). That is a separate issue to address.gpt-5.4in the docstring ofevals/improve.py:5was left as-is since it's documentation about what the loop is trying to do, not functional code.