Prune non-working Runware models with 0 calls

[devin-ai-integration]

Prune non-working Runware models with 0 calls

Summary

Removed 21 non-working Runware models that have 0 calls or "Coming soon" status on https://runware.ai/models. This cleanup reduces the model catalog from 143 to 122 models while keeping only verified working models.

Changes:

• 143 → 122 models (-21)
• Image: 88 → 85 models (-3)
• Video: 46 → 36 models (-10)
• Audio: 9 → 1 model (-8, kept only ElevenLabs Music V1 with 7.7K calls)

Removed models:

• Audio (8): All ElevenLabs TTS models with 0 calls (Flash v2/v2.5, Monolingual v1, Multilingual v1/v2, Turbo v2/v2.5, v3)
• Image (3): FLUX.1 [dev] SRPO, Kolors 2.0, Kolors 2.1
• Video (10): Google Veo 2, Veo 3.1, Sora 2, KlingAI 1.6 PRO/STD, MiniMax 01 Director, MiniMax 02 Hailuo, LTX-2 Fast, PixVerse v4.5, Vidu Q1

Review & Testing Checklist for Human

⚠️ HIGH PRIORITY - This PR removes models based on manual verification. Please verify:

• Verify removed models: Open https://runware.ai/models and confirm that all 21 removed models actually have 0 calls or "Coming soon" status. Check a few key ones like:

  • ElevenLabs audio models (should all be 0 except Music V1)
  • Google Veo 2 (google:2@0)
  • Sora 2 (openai:3@1)

• Test model selection UI:

  • Navigate through all model categories (image/video/audio)
  • Verify dropdowns load correctly
  • Try selecting different models and generating content
  • Check that no errors appear in console

• Verify ElevenLabs Music V1 still works: This is the only audio model kept. Test that:

  • It appears in the audio models dropdown
  • Music generation works with this model
  • The model card displays correctly

• Check for missing important models: Review the removed list above and confirm none of these are actually working models that should have been kept

• Test end-to-end workflow: Try a complete workflow (select model → configure → generate) for at least one model in each category to ensure nothing broke

Notes

• Verification was done manually by inspecting https://runware.ai/models on 2025-11-04
• The list of non-working AIR IDs was compiled based on screenshot and browser inspection
• Only lint checks were run programmatically; UI/runtime testing needs human verification
• User specifically noted that from audio models "похоже работает только одна" (only one seems to work)

Link to Devin run: https://app.devin.ai/sessions/5388bb693ed44ffb8aa537d7f4e7c79d
Requested by: Ilya Timonin (timoncool@gmail.com, @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 3:14am