[waleedlatif1]

Summary

• Expand the Video Generator’s searchable fal picker to 112 prompt-to-video endpoints. Keep saved model IDs working and hide the retired Sora entries from new selections.
• Show only settings each endpoint accepts, leave optional settings at the model default, and send durations and audio flags in the provider’s wire format.
• Validate queue URLs before attaching credentials, retain hosted receipts before media delivery, and distinguish video endpoints from image endpoints when falling back to estimated billing.

Type of Change

• New feature
• Bug fix

Testing

• Validated 1,808 block-to-queue request cases for all 112 endpoints against fal’s live OpenAPI schemas. Checked 52 existing block/provider configurations against staging and 88 duration/resolution combinations.
• Verified MP4 headers for 77 public examples and replayed a completed queue result through the production downloader. The resulting file decoded as H.264, 768×512, 5.04 seconds. Replayed all 112 models without optional queue links against the documented owner/app status and result routes. No paid inference jobs were submitted.
• Reproduced and fixed a Stop immediately after provider completion: metered calls retain the receipt for reconciliation without adding a second successful charge.
• Ran the video client, operation, hosted-key, and pricing tests, the repository test suite, lint, workspace type checks, all 62 audits, and the block registry checks.

Checklist

• Code follows project style guidelines
• Self-reviewed my changes
• Tests added/updated and passing
• No new warnings introduced
• I confirm that I have read and agree to the terms outlined in the Contributor License Agreement (CLA)

[vercel]

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

1 Skipped Deployment

Project	Deployment	Actions	Updated
docs	Skipped		Nov 3, 2026 10:06pm UTC

[greptile-apps]

Confidence Score: 5/5

[Medium impact] The changes appear safe to merge; no actionable new issue was established.

Summary

Expands the fal Video Generator catalog while keeping saved model IDs working.

• The Fal.ai picker lists searchable video models and their supported settings.
• Fal.ai video requests match each endpoint’s accepted settings and formats.
• Completed hosted video requests keep their cost receipts if media delivery stops.

No actionable new issues were established in the changes since the previous review.

_{Reviews (3) · Last reviewed commit: "fix(video): poll nested fal models throu..." · Reviewed by Greptile}

[cubic-dev-ai]

1 issue found across 10 files

Confidence score: 3/5

• In apps/sim/lib/internal/video/client.ts, a failed download or Stop can leave a Fal.ai charge without billing metadata. Reconcile the cost as soon as generation completes, independently of the download.

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="apps/sim/lib/internal/video/client.ts">

<violation number="1" location="apps/sim/lib/internal/video/client.ts:852">
P1: This downloads the completed Fal.ai media before reconciling hosted cost, so a download failure or Stop can leave a provider charge without billing metadata. Resolve the cost immediately after completion, independently of the caller abort signal, before downloading the media.

(Based on your team's feedback about completed Fal.ai cost reconciliation.)</violation>
</file>

<sub>Reply with feedback, questions, or to request a fix.

Turn on auto-fix | Re-trigger cubic</sub>

[waleedlatif1]

@greptile

[waleedlatif1]

@cubic-dev-ai review this PR

[cubic-dev-ai]

@cubic-dev-ai review this PR

@waleedlatif1 I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

1 issue found across 11 files

Confidence score: 3/5

• In apps/sim/lib/internal/video/client.ts, the fallback can return 405 for models with /text-to-video when convenience URLs are absent. Build the fallback from the owner/app segments instead.

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="apps/sim/lib/internal/video/client.ts">

<violation number="1" location="apps/sim/lib/internal/video/client.ts:784">
P2: The fallback uses the full model endpoint, but Fal request URLs use only the owner/app segments; models with `/text-to-video` can receive 405 responses when convenience URLs are absent. Build this fallback from `endpoint.split('/').slice(0, 2)` instead.</violation>
</file>

<sub>Reply with feedback, questions, or to request a fix.

Turn on auto-fix | Re-trigger cubic</sub>

[waleedlatif1]

@greptile

[waleedlatif1]

@cubic-dev-ai review this PR

[cubic-dev-ai]

@cubic-dev-ai review this PR

@waleedlatif1 I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

No issues found across 11 files

Confidence score: 5/5

• Automated review surfaced no issues in the provided summaries.
• No files require special attention.

_{Turn on auto-fix | Re-trigger cubic}