Image Generation Guide
Learn how to write effective prompts, choose dimensions, and get the best results from Rendra's AI image generator.
Prompt Best Practices
The quality of your generated image is directly tied to the quality of your prompt. Here are the key principles:
Be specific and descriptive
Include the visual style, color palette, typography feel, mood, and composition you want. Generic prompts produce generic results.
"A dark-themed SaaS blog OG image with a purple-to-indigo gradient background, bold white sans-serif title text on the left, and a subtle grid pattern overlay. Professional and modern."
"A blog header image."
Mention the format context
Tell the AI what the image will be used for — social card, blog OG, product announcement banner. This helps set the right composition and hierarchy.
Specify colors and style
Reference colors by name or hex, and describe the style: minimalist, bold, corporate, playful, dark/light. The AI uses these signals to set the overall aesthetic.
Include text placement hints
If you want text in the image, describe where it should sit: "title text centered at the top", "website URL in the bottom-right corner", "logo in the top-left".
Example Prompts
Supported Dimensions
Rendra generates images at any custom dimension, but these are the most common OG image sizes:
| Dimensions | Ratio | Use Case |
|---|---|---|
1200 × 630 | 1.91:1 | Facebook, LinkedIn, Open Graph standard |
1200 × 675 | 16:9 | Twitter/X large card, YouTube thumbnail |
1200 × 628 | ~1.91:1 | Facebook recommended |
1280 × 640 | 2:1 | Twitter summary card (legacy) |
800 × 418 | ~1.91:1 | Smaller OG, fast loading |
1200×630 — it works across all major social platforms and is the Open Graph standard.
Output Formats
PNG (default)
Lossless compression. Best for images with text, sharp edges, or transparency. Slightly larger file size than JPEG.
JPEG
Lossy compression. Smaller file sizes at the cost of some quality. Suitable for photographic or gradient-heavy images where small artifacts are acceptable.
Limiti di frequenza
Rate limits are applied per API key, per rolling 60-second window:
| Plan | Images / Month | Requests / Minute |
|---|---|---|
| Gratuito | 50 | 5 |
| Pro | 1,000 | 30 |
When you hit a rate limit you'll receive a 429 Too Many Requests response with a Retry-After header indicating how many seconds to wait.
See Fatturazione e utilizzo to upgrade your plan or check your current usage.