MxChat Video Generation

Generate stunning AI videos using Google’s Veo and OpenAI’s Sora models directly from your WordPress dashboard. Create high-quality videos from text prompts with advanced features like native audio generation, multiple resolutions and aspect ratios, flexible durations, and reference images for character consistency.

Requirements

  • MxChat Core Plugin (installed and activated)
  • MxChat Pro License (active)
  • Google Gemini API key for Veo models
  • OpenAI API key for Sora models (uses your existing MxChat OpenAI key)

Quick Setup

  1. Install the Plugin: Upload and activate MxChat Video Generation through your WordPress plugins page
  2. Get API Keys:
    • For Veo: Visit Google AI Studio to generate your Gemini API key
    • For Sora: Uses your existing OpenAI API key from MxChat settings
  3. Configure Settings: Navigate to MxChat → Video Generation and click the settings icon
  4. Enable & Add Key: Toggle “Enable Video Generation” and paste your Gemini API key (starts with “AIza…”)
  5. Test Generation: Try a simple prompt like “A cat playing piano” to verify everything works

Available Models

Google Veo Models

Veo 3.1 (Preview)

4-8 second videos with enhanced audio – Latest model with video extension and up to 3 reference images for style/character consistency. 720p/1080p at 24fps.

Generation time: 11 seconds to 6 minutes

Veo 3.1 Fast (Preview)

4-8 second videos with audio (speed optimized) – Fast generation with video extension support. 720p/1080p at 24fps.

Generation time: 8 seconds to 3 minutes

Veo 3

4-8 second videos with audio – Includes dialogue and sound effects. 720p/1080p at 24fps.

Generation time: 11 seconds to 6 minutes

Veo 3 Fast

4-8 second videos with audio (optimized) – Faster generation while maintaining quality. 720p/1080p at 24fps.

Generation time: 8 seconds to 3 minutes

Veo 2

5-8 second silent videos – Stable performance. 720p at 24fps.

Generation time: 15 seconds to 5 minutes

OpenAI Sora Models

Sora 2

4, 8, or 12 second videos with audio – Fast and flexible for rapid iteration. 720p/1080p at 24fps.

Generation time: 1 to 5 minutes

Sora 2 Pro

4, 8, or 12 second videos with audio – Production-quality output with higher fidelity for cinematic footage. 720p/1080p at 24fps.

Generation time: 2 to 10 minutes

Key Features

Reference Images

Upload a reference image to guide video generation and maintain character consistency. Veo 3.1 models support up to 3 reference images for enhanced style and character consistency throughout the video. Simply click “Choose Image” and select from your media library.

Negative Prompts (Veo only)

Specify what you don’t want in your video. Example: “low quality, blurry, distorted” helps generate cleaner results. Note: Sora models do not support negative prompts.

Resolutions

  • 720p: Available on all models
  • 1080p: Available on Veo 3.x and Sora models

Aspect Ratios

  • 16:9 Widescreen: Available on all models
  • 9:16 Portrait: Available on all models (perfect for social media)

Durations

  • Veo 3.x: 4, 6, or 8 seconds
  • Veo 2: 5, 6, or 8 seconds
  • Sora 2 / Sora 2 Pro: 4, 8, or 12 seconds

Audio Generation

Veo 3.x and Sora models generate videos with native audio, including dialogue and sound effects. Veo 2 generates silent videos.

Auto-save to Media Library

Generated videos are automatically saved to your WordPress media library with metadata including prompt and model used.

How to Generate Videos

  1. Write Your Prompt: Describe the video you want (max 1024 characters for Veo, 1000 for Sora)
  2. Select Model: Choose from Veo 3.1, Veo 3.1 Fast, Veo 3, Veo 3 Fast, Veo 2, Sora 2, or Sora 2 Pro
  3. Configure Settings: Select resolution, duration, and aspect ratio
  4. (Optional) Add Reference Image: Click “Choose Image” to select a reference for character consistency
  5. (Optional) Negative Prompt: Add negative prompts for Veo models
  6. Generate: Click “Generate Video” and wait for processing
  7. Save: Once complete, click “Save to Library” to store permanently
Pro Tip: Use keyboard shortcut Ctrl/Cmd + Enter while in the prompt field to quickly generate.

Effective Prompting Tips

Good Prompts Include:

  • Subject: Who or what is the focus
  • Action: What’s happening
  • Setting: Where it takes place
  • Style: Visual characteristics
  • Mood: Emotional tone

Example Prompts:

  • “A golden retriever running through autumn leaves in slow motion, cinematic lighting”
  • “Time-lapse of a blooming rose, macro shot, soft pink petals opening”
  • “Futuristic cityscape at sunset, flying cars, neon lights, cyberpunk style”

For Audio (Veo 3.x and Sora):

Include audio descriptions in your prompt: “A guitarist playing blues music, fingers strumming, authentic guitar sounds”

Troubleshooting

Error: “API key not configured”

For Veo: Go to Settings and ensure you’ve entered your Google Gemini API key correctly (starts with “AIza”).

For Sora: Ensure your OpenAI API key is configured in MxChat settings.

Error: “Generation failed” or timeout

  • Check your API key has the correct model access enabled
  • Verify you haven’t exceeded API quotas
  • Try a simpler prompt first
  • Generation can take up to 10 minutes for some models – be patient

Video won’t play or download

  • Ensure your browser supports MP4 video playback
  • Check your WordPress media upload limits
  • Try saving to media library first, then download

Reference image not working

  • Image must be under 20MB
  • Supported formats: JPEG, PNG, GIF, WebP
  • Try using a smaller or different image

Sora duration error

Sora models only support specific durations: 4, 8, or 12 seconds. Other values will result in an API error.

Frequently Asked Questions

Do I need MxChat Pro?

Yes, this add-on requires an active MxChat Pro license to function.

What API keys do I need?

For Veo models, you need a Google Gemini API key from Google AI Studio. For Sora models, it uses your existing OpenAI API key configured in MxChat settings.

What’s the video quality?

All models generate MP4 videos at 24fps. Resolution options include 720p (all models) and 1080p (Veo 3.x and Sora models).

What video durations are available?

Veo 3.x: 4, 6, or 8 seconds. Veo 2: 5, 6, or 8 seconds. Sora: 4, 8, or 12 seconds.

Can I generate multiple videos at once?

No, videos are generated one at a time. Wait for the current generation to complete before starting another.

Where are videos stored?

Videos are automatically saved to your WordPress media library under wp-content/uploads.

Is there a limit on generations?

Limits depend on your API quotas. Check your Google Cloud Console or OpenAI dashboard for usage details.

Can I use generated videos commercially?

Review Google’s and OpenAI’s terms of service for API usage. Generally, yes, but verify for your specific use case.

Which models support audio?

Veo 3.1, Veo 3.1 Fast, Veo 3, Veo 3 Fast, Sora 2, and Sora 2 Pro all support native audio generation. Veo 2 generates silent videos.

What’s the difference between Sora 2 and Sora 2 Pro?

Sora 2 is optimized for fast iteration and exploration. Sora 2 Pro offers production-quality output with higher fidelity for cinematic footage and marketing assets.

Common API Errors

Google Veo Errors

HTTP 403: Forbidden

Your API key doesn’t have permission to access Veo models. Check that Veo is enabled in your Google AI Studio project.

HTTP 429: Too Many Requests

You’ve hit the rate limit. Wait a few minutes before trying again or upgrade your API quota.

HTTP 400: Bad Request

Usually means invalid parameters. Check that your prompt isn’t too long (max 1024 chars) and doesn’t contain prohibited content.

OpenAI Sora Errors

HTTP 400: Invalid value for ‘seconds’

Sora only accepts 4, 8, or 12 seconds. Other values will be rejected.

HTTP 401: Unauthorized

Your OpenAI API key is invalid or expired. Check your API key in MxChat settings.

HTTP 500: Internal Server Error

The API is having issues. Wait a few minutes and try again.

Recent Updates

Version 1.0.5

Added OpenAI Sora 2 and Sora 2 Pro model support. Plugin renamed to “MxChat Video Generation”.

Version 1.0.4

Added Veo 3.1 and Veo 3.1 Fast preview models with video extension and multi-reference image support. Added resolution (720p/1080p) and duration selection. All models now support both aspect ratios.

Version 1.0.3

Added ability to generate videos with a reference image for character consistency.

Version 1.0.2

Fixed video history gallery. Added delete and download buttons to history items.

Version 1.0.1

Updated Google Veo 3 API integration. Added support for Veo 3 Fast model.

Need Help?

If you’re experiencing issues not covered here:

  • Visit MxChat.ai for additional resources
  • Check our YouTube channel for video tutorials
  • Contact MxChat support with your license key and error details