Text to Image

A new-generation image creation model ByteDance, Seedream 4.5 integrates image generation and image editing capabilities into a single, unified architecture.

Inference

Commercial use

Input

Form

Prompt*

A selfie of a cat, with the cat as the protagonist. The setting is twilight at the Eiffel Tower. The cat is happy, holding a piece of baklava in its paw. The photo has a slight motion blur and is slightly overexposed. From a selfie angle, with a bit of motion blur, the overall image presents a sense of calm madness. The text "Seedream 4.5 is on fal" should be written on the picture at the top in clearly visible font and crisp lettering. The image has a 4:3 aspect ratio

Additional Settings

More

Customize your input with more control.

Result

Idle

PreviewJSON

{
  "images": [
    {
      "url": "https://storage.googleapis.com/falserverless/example_outputs/seedreamv45/seedream_v45_t2i_output.png"
    }
  ],
  "seed": 42
}

Your request will cost $0.04 per image.

Logs

Seedream 4.5 [text-to-image]

ByteDance's Seedream 4.5 delivers up to 4 megapixels (2048x2048) at $0.04 per image, unifying text-to-image generation and image editing in a single architecture. Trading specialized endpoints for architectural efficiency, this model handles both creation and modification workflows without switching systems. Built for developers who need flexible image generation without managing multiple API integrations.

Built for: Product mockup generation | Marketing asset creation | Creative iteration workflows

Unified Architecture That Eliminates Endpoint Switching

Seedream 4.5 consolidates generation and editing capabilities into one model instead of forcing you to route between separate text-to-image and image-editing endpoints. This architectural decision means you can start with text prompts and move to refinement without changing API calls or managing different parameter schemas.

What this means for you:

• High-resolution output: Generate images up to 4 megapixels (2048x2048 maximum) with flexible aspect ratios between 1920px and 4096px per dimension
• Multi-image generation: Request up to 6 separate generations per API call, with the option to generate multiple variations per generation using max_images
• Deterministic control: Set seed values to reproduce exact outputs across runs, critical for A/B testing and iterative refinement
• Safety integration: Built-in content filtering enabled by default, with the option to disable for controlled environments

Performance That Scales

Seedream 4.5 positions at the accessible end of ByteDance's model family, prioritizing cost efficiency for high-volume workflows.

Technical Specifications

How It Stacks Up

Bytedance Seedream v4 Text to Image – Seedream 4.5 represents the latest iteration with architectural refinements for prompt interpretation. Seedream v4 remains available for workflows optimized around its specific parameter set.

Bytedance Text to Image (v3) – Seedream 4.5 trades the v3 generation's approach for unified editing capabilities at the same $0.04 price point. Version 3 prioritizes pure generation workflows without integrated editing support.

Emu 3.5 Image Text to Image – Seedream 4.5 offers unified generation-editing architecture for workflow consolidation. Emu 3.5 focuses on specialized generation with distinct architectural priorities for prompt interpretation.

Hunyuan Image Text to Image – Seedream 4.5 emphasizes architectural unification across generation and editing tasks. Hunyuan v3 prioritizes generation-specific optimizations for pure text-to-image workflows.