Update Runware models with extracted data from playground bundle

[devin-ai-integration]

Update Runware Models with Extracted Data from Playground Bundle

Summary

Extracted 132 models (79 image, 44 video, 9 audio) from the Runware playground JavaScript bundle and merged them with existing curated model data. The merge strategy preserves all manually-curated fields (dimensions, schedulers, min/max ranges) for existing models while updating labels, descriptions, and default values from the bundle. Added 53 new models with safe defaults only.

Key changes:

• Updated 79 existing models with refreshed metadata from Runware playground
• Added 53 new models (various image, video, and audio models)
• Preserved 11 old models not found in bundle
• Fixed category mapping: audio models now use "music" category (per ApiInfo type)
• Made description field optional in ApiInfo type to handle models without descriptions
• Created automation scripts for future model catalog updates

Review & Testing Checklist for Human

⚠️ CRITICAL - Data Quality Concerns: The user explicitly requested verification that extracted data won't break existing functionality. Please test thoroughly:

• Test existing popular models - Verify FLUX.1 [dev], Imagen 4.0, Bria FIBO, Riverflow models still generate correctly with updated default values
• Test new models - Try generating with 2-3 new models (e.g., Vidu Q1, KlingAI, MiniMax) to ensure they work end-to-end
• Test audio/music models - Verify ElevenLabs models work correctly with "music" category instead of "audio" (check if UI treats music vs voiceover differently)
• Verify UI stability - Check that model picker, parameter sliders, and generation UI work correctly with:
  • Models that have no description (should handle optional field gracefully)
  • New models that may have limited parameter ranges
  • Models with updated default steps/CFG values
• Spot-check defaults - Open a few models and verify that default values (steps, CFG, schedulers) seem reasonable and match playground behavior

Test Plan

1. Open the app and navigate to model picker
2. Select a previously working model (e.g., FLUX.1 [dev]) and generate an image - verify it works
3. Select a new model (e.g., Vidu Q1) and attempt generation - verify parameters look correct
4. Select an ElevenLabs music model and verify it appears in the correct category
5. Check that models without descriptions don't cause UI errors

Notes

• Risk: Unverified defaults - Default values (steps, CFG, schedulers) were extracted from JavaScript bundle but haven't been verified against actual Runware API behavior
• Risk: Category mapping - Changed audio → music, but TTS models might semantically be "voiceover" instead
• Risk: New models have limited metadata - 53 new models added with minimal fields to avoid fabricating ranges, but this means they may have less polished UI experience
• Automation ready - Created scripts in scripts/ directory for future model catalog updates

---

Session: https://app.devin.ai/sessions/5388bb693ed44ffb8aa537d7f4e7c79d
Requested by: Ilya Timonin (@timoncool)

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
videosos	Ready	Preview	Comment	Nov 4, 2025 2:54am