new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

AI SDK · Providers · all subjects

bytedance/capabilities

13 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

ByteDance image generation from text

Text-to-image generation is performed using `generateImage()` with `byteDance.image()` and a `prompt` string parameter.

ByteDance image editing via image-to-image

Image-to-image transformation is performed by passing `prompt.images` with a text description in `prompt.text`. Images may be provided as binary data, base64 strings, or URLs.

ByteDance multi-image blending

Multiple images can be blended to combine styles and elements from several references into a single output. Images are passed as an array via `prompt.images`.

ByteDance inpainting limitation

Seedream does not support mask-based inpainting. Edits must be described in the prompt, optionally with markers drawn on the input image (interactive editing, supported by `dola-seedream-5-0-pro-260628`).

ByteDance image streaming limitation

Streaming image output is not currently supported through the AI SDK integration.

ByteDance text-to-video generation

Text-to-video is performed using `experimental_generateVideo()` with `byteDance.video()`, a `prompt` string, `aspectRatio` (e.g. '16:9'), and `duration` (in seconds).

ByteDance image-to-video generation

Image-to-video generation uses a first-frame image provided via `prompt.image` with an optional text prompt in `prompt.text`. When the output ratio is dictated by input media, pass `aspectRatio: 'adaptive'` or omit `aspectRatio` entirely.

ByteDance audio-video synchronization

Seedance 1.5 Pro and Seedance 2.0 support generating synchronized audio alongside video via the `generateAudio` option in `providerOptions.bytedance`.

ByteDance first-and-last frame video generation

Generate smooth transitions between a starting and ending keyframe by providing `prompt.image` (first frame) and `providerOptions.bytedance.lastFrameImage` (last frame). Supported by Seedance 1.5 Pro, 1.0 Pro, and 1.0 Lite I2V.

ByteDance multi-reference image-to-video

Seedance 1.0 Lite I2V accepts 1-4 reference images via `providerOptions.bytedance.referenceImages` array. Use `[Image 1]`, `[Image 2]`, etc. in the prompt to reference specific images. The model extracts key features from each image and reproduces them in the video.

ByteDance reference video guidance

Seedance 2.0 supports reference videos (up to 3 videos, max 15 seconds each) via `providerOptions.bytedance.referenceVideos` array to guide the style, motion, or composition of the generated video.

ByteDance reference audio guidance

Seedance 2.0 supports reference audio (up to 3, max 15 seconds each) via `providerOptions.bytedance.referenceAudio` array for audio-guided video generation. The model uses referenced audio as background music or synchronized sound. Supports data URIs (e.g., `data:audio/wav;base64,...`).

ByteDance video output format

All ByteDance video models output MP4 video at 24 fps.

Give your agent this brain