feat(llm): add gpt-5.2-codex to verified models#1893
feat(llm): add gpt-5.2-codex to verified models#1893juanmichelini merged 1 commit intoOpenHands:mainfrom
Conversation
Add gpt-5.2-codex to both VERIFIED_OPENAI_MODELS and VERIFIED_OPENHANDS_MODELS to enable support in the OpenHands provider interface. Also add comprehensive tests for gpt-5.2-codex: - Prompt cache retention support - Reasoning effort support - Responses API support This completes the gpt-5.2 family support alongside the existing gpt-5.2 model. Related to OpenHands/OpenHands#12720 Co-authored-by: openhands <[email protected]>
all-hands-bot
left a comment
all-hands-bot
left a comment
There was a problem hiding this comment.
Code review complete. Found one ordering inconsistency between the two verified model lists that should be addressed for maintainability. Test coverage looks comprehensive.
| "gpt-5.2-codex", | ||
| "gpt-5.1-codex-max", | ||
| "gpt-5.1-codex", | ||
| "gpt-5.1", |
There was a problem hiding this comment.
🟠 Important: Model ordering is inconsistent between VERIFIED_OPENAI_MODELS and VERIFIED_OPENHANDS_MODELS.
In VERIFIED_OPENAI_MODELS (lines 2-6):
- gpt-5.2
- gpt-5.2-codex
- gpt-5.1 ← base model comes before codex variants
- gpt-5.1-codex-max
- gpt-5.1-codex
In VERIFIED_OPENHANDS_MODELS (lines 49-53):
- gpt-5.2
- gpt-5.2-codex
- gpt-5.1-codex-max ← codex variants come before base model
- gpt-5.1-codex
- gpt-5.1
This inconsistency could lead to maintenance confusion. Consider reordering VERIFIED_OPENHANDS_MODELS to match the pattern in VERIFIED_OPENAI_MODELS for consistency.
| @@ -1,5 +1,6 @@ | |||
| VERIFIED_OPENAI_MODELS = [ | |||
| "gpt-5.2", | |||
| "gpt-5.2-codex", | |||
There was a problem hiding this comment.
🟡 Suggestion: The gpt-5.1 family includes gpt-5.1-codex-max (line 5). Should the gpt-5.2 family also include gpt-5.2-codex-max for consistency?
If this variant does not exist in the OpenAI API, this is fine. Otherwise, consider adding it for completeness.
| ("gemini-3-flash-preview", True), | ||
| # GPT-5 family | ||
| ("gpt-5.2", True), | ||
| ("gpt-5.2-codex", True), |
There was a problem hiding this comment.
🟢 Nit: Excellent test coverage! The new model is tested across all three feature areas:
- Reasoning effort support (line 41)
- Responses API support (line 229)
- Extended retention support (line 264)
This ensures feature parity with gpt-5.2.
3 participants
Add gpt-5.2-codex to both VERIFIED_OPENAI_MODELS and VERIFIED_OPENHANDS_MODELS to enable support in the OpenHands provider interface.
Changes
gpt-5.2-codextoVERIFIED_OPENAI_MODELSgpt-5.2-codextoVERIFIED_OPENHANDS_MODELSThis completes the gpt-5.2 family support alongside the existing gpt-5.2 model.
Related to OpenHands/OpenHands#12720