Add GPT-5.5 support#143
Conversation
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: bb6cb2bc25
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| GPT_5_4_2026_03_05_MODEL: 272000, | ||
| GPT_5_4_MINI_2026_03_17_MODEL: 400000, | ||
| GPT_5_4_NANO_2026_03_17_MODEL: 400000, | ||
| GPT_5_5_2026_04_23_MODEL: 272000, |
There was a problem hiding this comment.
Use GPT-5.5 API context window for truncation limits
MODEL_CONTEXT_LENGTHS sets GPT_5_5_2026_04_23_MODEL to 272000, but GPT-5.5’s API context window is 1M tokens; this value is consumed by messages_within_context_window() to prune history, so any workspace using gpt-5.5 will drop a large portion of valid prompt context long before hitting the real model limit. That regression directly reduces answer quality for longer threads and defeats the main benefit of adding GPT-5.5 support.
Useful? React with 👍 / 👎.
There was a problem hiding this comment.
It's the pricing and scenario concern.
Ref: