fix: add support for max_completion_tokens parameter in newer OpenAI models#35

Open

caffeinum wants to merge 6 commits intofuergaosi233:mainfrom

caffeinum:fix/gpt5-max-completion-tokens

Commits on Aug 8, 2025

fix: add support for max_completion_tokens parameter in newer OpenAI models

caffeinum
and
claude
committed
fix: handle additional restrictions for newer OpenAI reasoning models

caffeinum
and
claude
committed
fix: disable streaming for reasoning models to avoid org verification errors

caffeinum
and
claude
committed
Revert "fix: disable streaming for reasoning models to avoid org verification errors"
caffeinum
committed
Revert "fix: handle additional restrictions for newer OpenAI reasoning models"
caffeinum
committed
fix: enforce temperature=1 for OpenAI reasoning models
caffeinum
committed