feat(together-ai): update model YAMLs [bot]

[harshiv-26]

Auto-generated by poc-agent for provider together-ai.

---

Note

Low Risk
Low risk because this only updates a provider model YAML (modalities/metadata) and does not change runtime logic; main risk is incorrect capability or pricing metadata affecting downstream routing/display.

Overview
Updates the Together AI vidu/vidu-q3 model YAML by adjusting the costs entry (including clarifying comments that pricing is not in official docs).

Expands declared modalities to accept image input and produce audio output, and adds sources plus status: active metadata.

^{Reviewed by Cursor Bugbot for commit bde36a1. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

/test-models

[harshiv-26]

Gateway test results

• Total: 1
• Passed: 0
• Failed: 0
• Validation failed: 0
• Errored: 0
• Skipped: 1
• Success rate: 0.0%

Provider	Model	Scenarios
together-ai	vidu/vidu-q3	skipped: skip-check

Skipped (1)

together-ai/vidu/vidu-q3 — skip-check (skipped)

Skip reason:

unsupported mode 'video'

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit bde36a1. Configure here.}

[cursor]

Incorrect audio output modality for video model

Medium Severity

The audio output modality was added to vidu/vidu-q3, but this is a video generation model (mode: video). No other video-mode model in the together-ai provider lists audio as an output — all sibling vidu models (vidu-q1, vidu-q3-turbo, vidu-2.0) and other video models (Seedance, Wan-AI) only output video. The only models with audio output are text-to-speech models. This will incorrectly advertise audio generation capabilities for this model.

 

^{Reviewed by Cursor Bugbot for commit bde36a1. Configure here.}