Generate Images from Text

​

Available Models

• higgsfield-ai/soul/standard - Our flagship text-to-image model
• reve/text-to-image - Versatile text-to-image generation
• And many more…

​

Quick Start

​

Basic Image Generation

curl -X POST 'https://platform.higgsfield.ai/higgsfield-ai/soul/standard' \
  --header 'Authorization: Key {your_api_key}:{your_api_key_secret}' \
  --header 'Content-Type: application/json' \
  --header 'Accept: application/json' \
  --data '{
    "prompt": "A serene mountain landscape at sunset with vibrant orange and purple skies",
    "aspect_ratio": "16:9",
    "resolution": "720p"
  }'

​

Best Practices

​

Writing Effective Prompts

• Be specific: Include details about style, mood, colors, and composition
• Use descriptive language: “A cozy mountain cabin at sunset with warm lighting” is better than “cabin”
• Specify artistic style: Mention styles like “photorealistic”, “watercolor”, “digital art”, etc.
• Include quality modifiers: Terms like “highly detailed”, “8k”, “professional” can improve results

​

Example Prompts

Good: "A professional photograph of a golden retriever puppy playing in a sunlit garden, shallow depth of field, bokeh background"

Better: "Professional pet photography of a golden retriever puppy, 3 months old, playing with a red ball in a lush green garden, golden hour lighting, shallow depth of field, Canon EOS R5, 85mm lens, bokeh background"

​

Optimizing Generation

• Start with lower resolutions for faster iterations
• Test different aspect ratios to find what works best for your use case
• Use webhooks for production workflows to avoid polling
• Store the request_id to retrieve results later if needed

​

Common Use Cases

• Marketing materials: Product images, social media content, advertisements
• Creative projects: Concept art, illustrations, design mockups
• Content creation: Blog headers, thumbnails, visual storytelling
• Prototyping: Quick visualization of ideas and concepts