feat: add support for max_completion_tokens parameter #45
+120
−2
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
Add support for the max_completion_tokens parameter required by newer OpenAI models (gpt-5, gpt-4o). This parameter is used instead of max_tokens for these models. Changes: - Added MaxCompletionTokens field to ChatCompletionNewParamsWrapper - Updated MarshalJSON to include max_completion_tokens in extras - Updated UnmarshalJSON to extract and set max_completion_tokens - Added comprehensive test coverage for the new parameter Fixes #40 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
There was a problem hiding this comment.
Pull Request Overview
This PR adds support for the max_completion_tokens parameter required by newer OpenAI models (gpt-4o, gpt-5) as an alternative to the older max_tokens parameter.
Key Changes:
- Added
MaxCompletionTokensfield toChatCompletionNewParamsWrapperstruct - Updated JSON marshaling/unmarshaling logic to handle the new parameter
- Added comprehensive test coverage for unmarshaling, marshaling, and nil handling scenarios
Reviewed Changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| openai.go | Adds MaxCompletionTokens field and updates marshal/unmarshal methods to handle the parameter |
| openai_test.go | Adds test suite covering unmarshal, marshal, and nil cases for the new parameter |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Address review feedback from GitHub Copilot. The previous implementation would skip setting max_completion_tokens when the value was 0, which is incorrect as 0 is a valid value that should be explicitly passed to the OpenAI API. Changes: - Check for field existence in JSON rather than value > 0 - Handle explicit 0 values correctly - Add test cases for zero value marshaling/unmarshaling - Support multiple numeric types (float64, int, int64) for robustness 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Revert the previous change that accepted zero values. Research shows that the OpenAI API requires positive integers for token limits, and there is no documentation supporting zero as a valid value. Changes: - Only accept max_completion_tokens > 0 - Add validation to reject 0 and negative values - Update tests to verify invalid values are ignored - Add test for negative value handling - Keep field existence check to improve over original implementation The GitHub Copilot review suggested accepting 0, but without any documentation proving it's valid, we should validate input to match API requirements. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
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
This PR adds support for the
max_completion_tokensparameter required by newer OpenAI models (gpt-5, gpt-4o) instead ofmax_tokens.Changes
MaxCompletionTokensfield toChatCompletionNewParamsWrapperMarshalJSONto includemax_completion_tokensin extras map when setUnmarshalJSONto extract and setmax_completion_tokensfrom incoming JSONTest Plan
TestMaxCompletionTokensverifies:make fmtRelated Issues
Fixes #40
🤖 Generated with Claude Code