feat(xai): add grok-imagine-video-1.5-preview

[analogpvt]

Summary

• Add xAI grok-imagine-video-1.5-preview video generation model
• $0.08/second, 720p/1080p, 5-15s duration
• Text-to-video and image-to-video support

Changes

• packages/models/src/models/xai.ts - Model definition
• apps/gateway/src/videos/videos.ts - Video job creation + proxy
• apps/worker/src/services/video-jobs.ts - Status polling base URL

🤖 Generated with Claude Code

Summary by CodeRabbit

• New Features
  • Integrated Reve provider for image generation with configurable aspect ratios
  • Added xAI provider support for video generation with advanced video format options
  • Introduced new xAI video model (grok-imagine-video-1.5-preview) featuring flexible video sizes, durations, and pricing tiers

[coderabbitai]

Caution

Review failed

Pull request was closed or merged during review

Walkthrough

This PR adds support for two new LLM gateway providers: Reve (image generation) and extends xAI with video generation. Reve is integrated with full request/response handling and model registry entries, while xAI's video capability is exposed through a new upstream job creator and model definition.

Changes

Reve Image Generation Provider

Layer / File(s)	Summary
Provider registration and model definitions packages/models/src/providers.ts, packages/models/src/models/reve.ts, packages/models/src/models.ts, .env.example	Reve provider is registered with API key configuration (LLM_REVE_API_KEY), a single "reve-create" image model is defined with pricing and feature flags, and both are wired into the global provider and model registries.
Request endpoint and body preparation packages/actions/src/get-provider-endpoint.ts, packages/actions/src/prepare-request-body.ts	Reve requests are routed to https://api.reve.com and mapped to the /v1/image/create endpoint; request bodies extract prompts and reference images from user messages and build a Reve-shaped request with version/aspect ratio metadata.
Response parsing and OpenAI transformation apps/gateway/src/chat/tools/parse-provider-response.ts, apps/gateway/src/chat/tools/transform-response-to-openai.ts	Reve's base64 image responses are recognized and parsed into images arrays; non-OpenAI-shaped responses with images are transformed into OpenAI-compatible chat completions with usage and metadata.

xAI Video Generation Provider

Layer / File(s)	Summary
Video model definition packages/models/src/models/xai.ts	The grok-imagine-video-1-5-preview video model is added to xAI's catalog with pricing (input/output/per-second rates), supported video sizes (720p/1080p), duration limits (5–60 seconds), and feature flags.
Video job creation and routing apps/gateway/src/videos/videos.ts, apps/worker/src/services/video-jobs.ts	xAI is added as a default provider base URL option, included in upstream proxy routing logic, and a new createXaiVideoJob function submits requests to /v1/videos/generations with optional first-frame images; the worker layer recognizes xAI and returns its API base URL.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Possibly related PRs

• theopenco/llmgateway#2059: Both PRs extend image-generation request/response plumbing in parse-provider-response.ts and transform-response-to-openai.ts to handle provider-specific (non-OpenAI-shaped) image outputs.
• theopenco/llmgateway#2379: Both PRs extend packages/models/src/models/xai.ts by adding new xAI model definitions to the exported array.
• theopenco/llmgateway#2133: Both PRs modify packages/models/src/models/xai.ts to extend the xaiModels array with new model entries.

Suggested reviewers

• steebchen
• smakosh

🚥 Pre-merge checks | ✅ 3 | ❌ 2

❌ Failed checks (2 warnings)

Check name	Status	Explanation	Resolution
Title check	⚠️ Warning	The PR title 'feat(xai): add grok-imagine-video-1.5-preview' accurately describes the primary change—adding a new xAI video generation model. However, the changeset also includes substantial support for a new Reve image generation provider with multiple file changes across the codebase, which is not mentioned in the title.	Consider updating the title to reflect both major additions, such as 'feat: add xai video model and reve image provider' or split into separate PRs, as the current title only partially represents the full scope of changes.
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}