# Seedream 5.0 Lite

Seedream 5.0 Lite adds web-connected retrieval to the Seedream image stack. The model can pull current information from the web during generation, so outputs reflect real-world context rather than training-cutoff knowledge alone.

- **Model ID:** `bytedance/seedream-5.0-lite`
- **Type:** image
- **Providers:** bytedance
- **Context window:** Not applicable
- **Maximum output tokens:** 0
- **Pricing:** $0.035/image
- **Canonical page:** https://vercel.com/ai-gateway/models/seedream-5.0-lite

## Supported parameters

- **Parameters:** `aspectRatio`, `resolution`
- **Operations:** `text-to-image`, `image-to-image`
- **Input modalities:** `text`, `image`
- **Output modalities:** `image`
- **Aspect ratios:** `1:1`, `16:9`, `9:16`, `4:3`, `3:4`, `3:2`, `2:3`
- **Resolutions:** `2K`, `3K`
- **Durations:** Not reported
- **Frame rates:** Not reported
- **Maximum outputs per request:** Not reported
- **Maximum input images:** Not reported
- **Input image formats:** `png`, `jpeg`
- **Maximum input image size:** 10 MB

## Example

```ts
import { experimental_generateImage as generateImage } from 'ai';

const result = await generateImage({
  model: 'bytedance/seedream-5.0-lite',
  prompt: 'A red balloon on a wooden table.'
});
```

## About

Seedream 5.0 Lite is the first Seedream model to introduce web-connected retrieval. During generation, the model can fetch real-time information from the web and fuse it with its multimodal generation capabilities. This is useful when the visual output needs to align with current facts, recent events, or evolving subject matter that a training-cutoff model would miss.

The Lite designation reflects positioning rather than a stripped feature set. Seedream 5.0 Lite suits production workflows that want web-grounded image generation without the cost profile of a flagship release. ByteDance highlights upgraded reasoning, broader global knowledge coverage, stronger reference consistency, and improved professional scene generation tuned for enterprise workflows.

Like other Seedream models, Seedream 5.0 Lite supports multimodal inputs. You can combine text prompts with reference images, and the retrieval system layers on top of that input rather than replacing it. The result is image generation that can reason over both the prompt and live context retrieved during the request.

Seedream 5.0 Lite is a dedicated image generation model. Call it through the AI SDK's `generateImage` function with the model set to `bytedance/seedream-5.0-lite`. AI Gateway handles authentication and provider routing.

## What to consider

Web-connected retrieval changes the latency and determinism profile compared with offline image generation. If your application requires fully deterministic output for the same prompt, factor in the variability that real-time retrieval introduces. See $0.035 for per-image pricing and N/A for the full breakdown.

## When to use

### Best For

- Image generation tied to current events, recent product launches, or evolving subject matter where training-cutoff knowledge falls short
- Enterprise content workflows that benefit from broader global knowledge coverage and stronger reference consistency
- Professional scene generation where the visual brief depends on accurate, current context
- Multimodal pipelines that need web-grounded reasoning layered on top of text and reference image inputs

### Consider Alternatives When

- Determinism matters more than current context. Seedream 4.0 or Seedream 4.5 produce more reproducible outputs for the same prompt
- Editing consistency is the primary requirement. Seedream 4.5 targets that workload directly
- Generation latency must stay minimal. Real-time retrieval adds variability to per-request timing

## Best for

- **Current-Context Generation:** Web-connected retrieval grounds outputs in real-time information beyond the training cutoff
- **Enterprise Scene Generation:** Stronger reference consistency and global knowledge coverage suit professional workflows
- **Multimodal Reasoning:** Text, reference images, and live web context combine in one generation step
- **Knowledge-Heavy Briefs:** Visual outputs that depend on accurate, evolving subject matter benefit from retrieval grounding

## Consider alternatives

- **Reproducible Output:** Seedream 4.0 or Seedream 4.5 produce more deterministic results for the same prompt
- **Editing Consistency:** Seedream 4.5 targets portrait, small-text, and color-tone preservation directly
- **Minimal Latency:** Real-time retrieval adds variability that purely offline generation avoids

## Frequently asked questions

### What does web-connected retrieval do in Seedream 5.0 Lite?

It lets the model fetch real-time information from the web during generation and fuse that information with its multimodal output. This grounds image generation in current context rather than relying only on what the model learned at training time.

### How does Seedream 5.0 Lite compare to Seedream 4.5?

Seedream 4.5 focuses on editing consistency, stylization, and prompt adherence. Seedream 5.0 Lite introduces web-connected retrieval and emphasizes upgraded reasoning, broader knowledge coverage, and professional scene generation. Pick based on whether retrieval grounding or editing fidelity matters more for your workflow.

### Does Seedream 5.0 Lite accept reference images?

Yes. Seedream 5.0 Lite supports multimodal inputs, so you can combine a text prompt with reference images. The retrieval system layers on top of those inputs rather than replacing them.

### How do I call Seedream 5.0 Lite through AI Gateway?

Set the model to `bytedance/seedream-5.0-lite` and call it through the AI SDK's `generateImage` function. AI Gateway routes the request through `bytedance` using your API key or OIDC token.

### What does Seedream 5.0 Lite cost per image?

Seedream 5.0 Lite is listed at $0.035 per image. See N/A for the full pricing breakdown.

### Does Vercel AI Gateway support Zero Data Retention for Seedream 5.0 Lite?

Zero Data Retention is not currently available for this model. Zero Data Retention is offered on a per-provider basis. See https://vercel.com/docs/ai-gateway/capabilities/zdr for details.

## Links

- [Model page](https://vercel.com/ai-gateway/models/seedream-5.0-lite)
- [AI Gateway documentation](https://vercel.com/docs/ai-gateway)
- [Provider model documentation](https://docs.byteplus.com/en/docs)
- [Provider pricing](https://docs.byteplus.com/en/docs/ModelArk/1544106)


If you encounter incorrect information, can’t find what you need, need a capability Vercel doesn’t currently offer, or have any other feedback to report, please do it by following the instructions at https://vercel.com/feedback.md.