Socializioz includes AI tools for generating text, images, and videos. You can access these from the media library, the composer, or through the AI chat assistant.Documentation Index
Fetch the complete documentation index at: https://docs.socializioz.com/llms.txt
Use this file to discover all available pages before exploring further.
Caption generation
Generate captions tailored to a specific platform and tone.Open the composer
Navigate to the composer and start a new post.
Click AI Write
In the editor toolbar, click the AI Write button to open the caption generation panel.
Configure options
Select the target platform, tone (Professional, Casual, or Friendly), and length (Short, Medium, or Long). The guided content builder offers two additional tones — Funny and Inspiring — in its advanced caption settings.
Caption enhancement
If you already have a draft caption, use the Enhance tool in the composer toolbar to improve it. The AI refines your wording, grammar, and tone while preserving your original idea. This is different from caption generation, which creates a new caption from scratch.Prompt enhancement
The AI content generation panel includes an Enhance button next to the prompt field. When you click it, the AI rewrites your simple description into a detailed, platform-optimized prompt. The enhancement considers:- Your target platform — Instagram, Facebook, TikTok, X (Twitter), LinkedIn, or Telegram
- Content language
- Industry context from your workspace settings
This is different from the Enhance tool in the composer, which improves existing caption text. The prompt enhancement here optimizes your AI generation prompts.
Image generation
Create images from a text prompt with options for aspect ratio and visual style.Open the Generate tab
Go to the media library and switch to the Generate tab, or use the
/image command in AI chat.Write a prompt
Describe the image you want. Click Enhance to let the AI expand your description into a detailed prompt.
Aspect ratios
| Ratio | Orientation | Best for |
|---|---|---|
| 1:1 | Square | Instagram feed, general use |
| 4:5 | Portrait | Instagram feed |
| 16:9 | Landscape | Facebook, LinkedIn, X |
Style presets
- Realistic — photographic style
- Illustration — drawn or illustrated look
- Minimal — clean, simple design
- Cinematic — dramatic, film-like aesthetic
Advanced image options
Expand the Advanced section below the style presets to access additional controls:- Industry — enter your industry (for example, “Car rental”) to give the AI more context about your brand when generating images. This helps produce visuals that are relevant to your sector.
- City — enter a city name (for example, “Dubai”) to add location context. The AI uses this to tailor imagery with appropriate landmarks, architecture, or cultural elements.
- Auto caption — enabled by default. When turned on, the AI automatically generates a caption alongside your image. Toggle this off if you only want the image without a caption.
Video generation
Generate short videos from text prompts or reference images. Socializioz supports two video generation providers.Open the Generate tab
Go to the media library and switch to the Generate tab.
Write a prompt
Describe the video you want. For image-to-video models like Gen-4.5, attach a reference image. For Hedra character models, configure voice and text-to-speech settings.
Choose a provider and model
Expand the Advanced options panel to select your video provider (Runway or Hedra) and model. Hedra is the default. Each model shows its supported resolutions, durations, and a short description of its best use case. Models that require an input image (such as image-to-video) are hidden when you are using a text-only prompt. See choosing a provider below for guidance.
Set duration and resolution
Choose the video duration and resolution based on the selected model’s supported options.
Choosing a video provider
Runway and Hedra serve different use cases. Use this guide to pick the right provider for your content.| Factor | Runway | Hedra |
|---|---|---|
| Best for | Scene-based videos, product showcases, cinematic clips | Talking-head videos, character animations, voice narration |
| Input types | Text prompts (Veo models) or reference image (Gen-4.5) | Text prompts, reference images, and keyframe pairs |
| Audio | AI-generated audio on Veo models only | Voice synthesis via ElevenLabs on character models |
| Max resolution | Up to 1920x1080 (Veo) or 1280x720 (Gen-4.5) | Varies by model — check the model selector |
| Speed | Veo 3.1 Fast is the quickest option for drafts | Processing time varies by model and duration |
| Credits | Counts toward your plan’s monthly AI videos limit | Uses Hedra credits (separate balance) plus your plan’s video limit |
| Vertical video | Gen-4.5 supports 720x1280 for Stories and Reels | Configurable aspect ratio per model |
When to use Runway
- You need a polished, cinematic-quality clip for a product launch or brand video
- You have a reference image and want to animate it into a short video (Gen-4.5)
- You want AI-generated audio paired with the video (Veo models)
- You need to iterate quickly on drafts (Veo 3.1 Fast)
When to use Hedra
- You want a spokesperson or character to deliver a message on camera
- You need voice narration using text-to-speech (ElevenLabs integration)
- You want to control the start and end frames of the video using keyframes
- You are producing social announcements, product walkthroughs, or educational content where a speaking presenter drives engagement
Runway models
| Model | Type | Resolutions | Duration | Audio | Best for |
|---|---|---|---|---|---|
| Gen-4.5 | Image to video | 1280x720, 720x1280 | 2-10 sec | No | General social videos, balanced quality and speed |
| Veo 3.1 | Text to video | Up to 1920x1080 | 4, 6, 8 sec | Yes | High visual detail |
| Veo 3.1 Fast | Text to video | Up to 1920x1080 | 4, 6, 8 sec | Yes | Quick drafts and iteration |
| Veo 3 | Text to video | Up to 1920x1080 | 4, 6, 8 sec | Yes | Reliable baseline quality |
Hedra
Hedra is an alternative video generation provider with support for character-based videos and voice synthesis.- Text-to-video — generate videos from a text prompt with configurable aspect ratio, resolution, and duration
- Character models — create talking-head style videos using ElevenLabs text-to-speech with selectable voice options
- Keyframe support — use start and end reference images to guide the video generation
Using character models with voice synthesis
Character models produce talking-head style videos where an AI-generated character speaks your script. When you select a character model:- Enter the text you want the character to say in the prompt field.
- Choose a voice from the ElevenLabs voice selector. Options include different genders, accents, and speaking styles.
- Configure the aspect ratio and duration, then generate.
Using keyframes
Keyframes let you control the start and end states of a generated video by providing reference images. This gives you more creative control compared to text-only prompts.- Start keyframe — an image that defines what the opening frame of the video looks like
- End keyframe — an image that defines the closing frame
Not all Hedra models support keyframes. When a model does not support keyframes, the keyframe upload fields are hidden. Check the model description in the selector to confirm keyframe compatibility.
- Character models — designed for talking-head and character-driven videos with voice synthesis
- Other models — general-purpose video generation models
| Plan | Monthly video credits |
|---|---|
| Free | 100 |
| Starter | 500 |
| Professional | 2,000 |
Create a video for social media
This workflow walks you through generating a video and publishing it to a platform.Decide on the video type
Determine whether you need a scene-based clip (use Runway) or a talking-head video with narration (use Hedra). See choosing a provider above.
Write and enhance your prompt
Open the Generate tab in the media library and select Video. Write a clear description of the video you want. Click Enhance to let the AI expand your prompt into a detailed, platform-optimized description.
Configure provider settings
Expand Advanced options and select your provider and model. For Hedra character models, choose a voice from the ElevenLabs selector. For Runway Gen-4.5, attach a reference image. Set the aspect ratio to match your target platform — 9:16 for TikTok, Instagram Reels, and Stories, or 16:9 for Facebook and LinkedIn.
Generate and wait
Click Generate. Video generation takes longer than images. The asset appears in your library with a Processing status. The library grid refreshes automatically every 12 seconds, so you can continue working while it processes.
Review the result
Once the status changes to Ready, click the video to preview it. Check the video resolution badge to confirm it meets your target platform’s requirements.
Use in a post
Click Use in post on the generated asset, or open the composer and select the video from your media library. The composer’s resolution hints confirm whether the video meets platform specs before you publish.
AI campaign planner
The AI can generate a full content calendar for a campaign, including post ideas, captions, media prompts, hashtags, and SEO keywords. You can access the planner from the analytics dashboard on Professional and Enterprise plans, or ask the assistant to plan a campaign in chat. See AI chat assistant — AI campaign planner for the full list of inputs and outputs.Usage limits
AI generation counts toward your plan’s monthly limits:| Resource | Free | Starter | Professional | Enterprise |
|---|---|---|---|---|
| AI images | 5/month | 50/month | 200/month | Unlimited |
| AI videos | 1/month | 15/month | 50/month | Unlimited |
| AI reels | 1/month | 15/month | 50/month | Unlimited |
Hedra credits
Hedra uses its own credit-based system on top of your plan’s monthly video limits. Your remaining Hedra credit balance is displayed in the top-right corner of the Hedra provider card. Each model shows its credit cost before you generate. When credits drop below 50, the balance turns red as a warning.| Plan | Monthly Hedra credits |
|---|---|
| Free | 100 |
| Starter | 500 |
| Professional | 2,000 |
Enterprise fair usage
Enterprise plans have unlimited monthly generation limits, subject to a fair usage policy: up to 100 AI images, 50 AI videos, and 50 AI reels per day per user. Contact the team for higher daily throughput.Troubleshooting
| Problem | Resolution |
|---|---|
| Generation fails or times out | Check the system status page. AI services may be temporarily unavailable. Try again in a few minutes. |
| Image does not match prompt | Use the Enhance button to create a more detailed prompt. Adding specific details about composition, lighting, and subject produces better results. |
| Video stuck in Processing | Video generation can take several minutes. Refresh the page and check the media library status filter. If it stays stuck for more than 10 minutes, try generating again. |
| ”Limit reached” error | You have used all of your plan’s monthly AI generation credits. Check your usage in Settings > Billing or upgrade your plan. |
| Generated caption does not match brand | Configure your brand profile with tone, audience, and example text so the AI can match your voice. |
| Video resolution does not meet platform requirements | Check the video resolution reference for your target platform’s specs before generating. Choose a model and resolution that match — for example, 720x1280 on Gen-4.5 for vertical TikTok or Reels content. |
| Hedra credits ran out but plan limit remains | Hedra credits and your plan’s monthly video limit are separate counters. Both must have remaining capacity. Check your Hedra balance in the provider card and your plan usage in Settings > Billing. |
| Character model video has no voice | Make sure you selected a voice from the ElevenLabs voice selector. Character models require a voice selection to produce spoken narration. If no voices appear, try refreshing the page. |
| Gen-4.5 model is not available | Gen-4.5 is an image-to-video model and only appears when you have attached a reference image. Switch to text-only prompt mode to see Veo and Hedra text-to-video models instead. |