fix: set supports_vision to false for novita/openai/gpt-oss-120b

[wasnertobias]

Why are these changes needed?

The novita/openai/gpt-oss-120b entry had supports_vision: true, but gpt-oss-120b is a text-only model without vision support.

Fixes #37584

Change type

• Bugfix
• New feature
• Refactor
• Documentation
• Performance improvement
• Test update
• CI/CD changes
• Other

Checklists

I submitted a PR to use-prompt repo

• I submitted the use-prompt update in a separate PR
• Not applicable

Model (if applicable)

• Model is already in the repo under the right provider folder, if not, I submitted a PR to litellm with the following changes:
  • Model added to model_prices_and_context_window_backup.json
  • Model added to litellm/model_prices_and_context_window_backup.json
  • Model added to relevant provider_list in schema.prisma
  • Model added to litellm/llms/<provider>/chat/transformation.py
  • Model added to litellm/llms/<provider>/<sub-provider>/chat/transformation.py if needed
  • Model added to relevant test in tests_llm_completion
• Model is already in the repo and I submitted a PR to litellm with the following changes:
  • Model added to model_prices_and_context_window_backup.json
  • Model added to litellm/model_prices_and_context_window_backup.json
• Not applicable

Billed as tokens (if applicable)

Billed as characters (if applicable)

Billed per request (if applicable)

Billed per image (if applicable)

Billed per audio (if applicable)

Billed per second (if applicable)

Related Issues

• Fixes Fix "novita/openai/gpt-oss-120b" entry in "model_prices_and_context_window.json" #37584

[CLAassistant]

All committers have signed the CLA.

[greptile-apps]

Greptile Summary

Corrects the Novita gpt-oss-120b capability metadata to identify the model as text-only.

• Sets supports_vision to false in the canonical model registry.
• Applies the same correction to the packaged backup registry.

Confidence Score: 5/5

The PR appears safe to merge because both model registries consistently mark the text-only model as not supporting vision.

The capability correction is applied to the exact Novita model entry in both synchronized registry copies, including the backup consumed by the packaged runtime.

Important Files Changed

Filename	Overview
model_prices_and_context_window.json	Correctly disables vision support for novita/openai/gpt-oss-120b in the canonical model registry.
litellm/model_prices_and_context_window_backup.json	Mirrors the capability correction in the packaged runtime registry, preserving exact synchronization with the canonical file.

_{Reviews (1): Last reviewed commit: "fix: set supports_vision to false for no..." | Re-trigger Greptile}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[codspeed-hq]

Merging this PR will not alter performance

✅ 31 untouched benchmarks

---

_{Comparing wasnertobias:fix/novita-gpt-oss-120b-supports-vision (8a8d608) with litellm_internal_staging (47a7e17)¹}

Footnotes

1. No successful run was found on litellm_internal_staging (5290150) during the generation of this report, so 47a7e17 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩