# Seedream 4

> Seedream 4 is available on Artifio.ai — a unified AI workspace with a single pay-as-you-go wallet.

- Canonical page: https://artifio.ai/models/bytedance-seedream-v4-edit
- Run it: https://artifio.ai/create/bytedance-seedream-v4-edit
- Modality: image editing
- Model brand: bytedance
- Pricing: from about $0.025 per generation (exact cost shown in the workspace before you confirm; varies with selected options)
- Typical generation time: ~40s (catalogue estimate)
- Full pricing table: https://artifio.ai/pricing/models

## About Seedream 4

Seedream 4.0 is ByteDance Seed's unified image model, released September 9, 2025, and this variant is its editing mode: you supply an existing image, or several, plus a plain-language instruction, and the model rewrites the picture. Because generation and editing share one architecture, edits inherit the same diffusion transformer speed work that gives the family its fast inference.

ByteDance reported that Seedream 4.0 ranked first for single-image editing on its MagicArena Elo ratings at launch, and the technical report notes a first-place debut on the Artificial Analysis image-editing leaderboard in September 2025. On Artifio it sits in the same workspace as 100+ other models, pay-per-generation rather than per-model subscription.

## Features

- Prompt-based edits that add, remove, modify, or replace elements described in natural language, with no masks or selection tools.
- Background replacement and portrait retouching handled as single instructions.
- Accepts up to a dozen reference images in one request for multi-image editing and composition.
- Multi-image output, useful for producing several edited variations in a batch.
- Can work from a line draft, turning a rough sketch plus references into a finished scene.
- Edited results render at up to 4K resolution.

## What people use it for

- Product photo cleanup, such as removing stray objects or swapping a distracting background.
- Changing or correcting text on signage, packaging, and posters inside a photo.
- Combining subjects or elements from several photos into one composed image.
- Portrait retouching without manual masking.
- Turning line drafts or sketches into rendered artwork.

## Reported strengths

- Instruction-following edits are the model's core design goal, so common operations need one sentence instead of layer work.
- Multi-image composition with up to a dozen references was unusual capacity at release.
- Still competitive on public arenas long after launch: rank 22 with an Elo of 1183 on the Artificial Analysis image-editing leaderboard as of July 2026.
- The unified architecture means edits keep the same text-rendering quality as fresh generations.

## Reported limitations

- GPT Image 2, GPT Image 1.5, Nano Banana 2, and MAI-Image-2.5 all rate above it on the Artificial Analysis image-editing arena as of July 2026.
- ByteDance's own Seedream 4.5 notes call out detail preservation in multi-image editing as an area that version improved, which implies 4.0 can drift on facial features, lighting, and color tone when several references are involved.
- The first-place editing claim at launch came from MagicArena, an evaluation platform ByteDance operates itself.

## Sources

1. Seedream 4.0 Officially Released: Beyond Drawing, Into Imagination (ByteDance Seed): https://seed.bytedance.com/en/blog/seedream-4-0-officially-released-beyond-drawing-into-imagination
2. Seedream 4.0: Toward Next-generation Multimodal Image Generation (arXiv): https://arxiv.org/abs/2509.20427
3. Seedream 4.0 (ByteDance Seed): https://seed.bytedance.com/en/seedream4_0
4. Image Editing Leaderboard (Artificial Analysis): https://artificialanalysis.ai/image/leaderboard/editing

---

About Artifio.ai: 100+ AI image, video, and audio models in one workspace with a single shared wallet — no per-model subscriptions. This is the markdown rendition of https://artifio.ai/models/bytedance-seedream-v4-edit for AI assistants and crawlers; the HTML page is canonical.