Getting Started

Welcome to Seedance 20 — your direct line to ByteDance's cinematic AI video generator. Describe a scene, paste an image, or tell me the mood you're going for, and I'll generate a high-quality video clip for you right now.

Try saying:

• "Animate this image into a video"
• "Generate a cinematic nature clip"
• "Create a product showcase video"

Automatic Setup

On first interaction, connect to the processing API before doing anything else. Show a brief status like "Setting things up...".

Token: If NEMO_TOKEN environment variable is already set, use it and skip to Session below.

Free token: Generate a UUID as client identifier, then POST to https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token with header X-Client-Id: <uuid>. The response field data.token becomes your NEMO_TOKEN (100 credits, 7-day expiry).

Session: POST to https://mega-api-prod.nemovideo.ai/api/tasks/me/with-session/nemo_agent with Bearer auth and body {"task_name":"project"}. Save session_id from the response.

Confirm to the user you're connected and ready. Don't print tokens or raw JSON.

Generate Stunning Videos in Seconds with Seedance 20

Seedance-20 is a state-of-the-art AI video generation model developed by ByteDance, designed to turn written descriptions and still images into fluid, high-resolution video clips. Whether you're visualizing a product concept, crafting a social media reel, or building a storyboard, seedance-20 handles the heavy lifting so you can focus on the creative vision.

Unlike older generation models that struggle with motion blur, inconsistent subjects, or unnatural transitions, seedance-20 produces videos with smooth temporal coherence — meaning objects move naturally and scenes flow the way a real camera would capture them. You can generate from pure text, use an image as a visual anchor, or combine both for precise control over the output.

This skill gives you direct access to seedance-20's capabilities in a conversational interface. Describe what you want, upload a reference image if needed, and receive a ready-to-use video clip. It's the fastest path from concept to content, no timeline editor or rendering queue required.

Routing Text and Image Prompts

When you submit a request, Seedance 20 automatically detects whether you're providing a text prompt, a reference image, or both, and routes your input to the appropriate generation pipeline for cinematic output.

Seedance 20 API Reference

Seedance 20 runs on a distributed cloud inference backend optimized for high-fidelity temporal consistency and motion coherence across frames. Video generation jobs are queued, rendered asynchronously, and returned as downloadable MP4 assets once processing completes.

Skill attribution — read from this file's YAML frontmatter at runtime:

• X-Skill-Source: seedance-20
• X-Skill-Version: from frontmatter version
• X-Skill-Platform: detect from install path (~/.clawhub/ → clawhub, ~/.cursor/skills/ → cursor, else unknown)

All requests must include: Authorization: Bearer <NEMO_TOKEN>, X-Skill-Source, X-Skill-Version, X-Skill-Platform. Missing attribution headers will cause export to fail with 402.

API base: https://mega-api-prod.nemovideo.ai

Create session: POST /api/tasks/me/with-session/nemo_agent — body {"task_name":"project","language":"<lang>"} — returns task_id, session_id.

Send message (SSE): POST /run_sse — body {"app_name":"nemo_agent","user_id":"me","session_id":"<sid>","new_message":{"parts":[{"text":"<msg>"}]}} with Accept: text/event-stream. Max timeout: 15 minutes.

Upload: POST /api/upload-video/nemo_agent/me/<sid> — file: multipart -F "files=@/path", or URL: {"urls":["<url>"],"source_type":"url"}

Credits: GET /api/credits/balance/simple — returns available, frozen, total

Session state: GET /api/state/nemo_agent/me/<sid>/latest — key fields: data.state.draft, data.state.video_infos, data.state.generated_media

Export (free, no credits): POST /api/render/proxy/lambda — body {"id":"render_<ts>","sessionId":"<sid>","draft":<json>,"output":{"format":"mp4","quality":"high"}}. Poll GET /api/render/proxy/lambda/<id> every 30s until status = completed. Download URL at output.url.

Supported formats: mp4, mov, avi, webm, mkv, jpg, png, gif, webp, mp3, wav, m4a, aac.

SSE Event Handling

~30% of editing operations return no text in the SSE stream. When this happens: poll session state to verify the edit was applied, then summarize changes to the user.

Backend Response Translation

The backend assumes a GUI exists. Translate these into API actions:

Draft field mapping: t=tracks, tt=track type (0=video, 1=audio, 7=text), sg=segments, d=duration(ms), m=metadata.

Timeline (3 tracks): 1. Video: city timelapse (0-10s) 2. BGM: Lo-fi (0-10s, 35%) 3. Title: "Urban Dreams" (0-3s)

Error Handling

Common Workflows

The most popular workflow with seedance-20 starts with a detailed text prompt. Users describe the subject, environment, lighting, camera movement, and mood — the more specific, the better the output. For example, specifying 'handheld camera, golden hour, shallow depth of field' produces noticeably different results than a bare subject description.

A second common workflow is image-to-video generation. You upload a still image — a product photo, a portrait, a landscape shot — and seedance-20 animates it based on your direction. This is particularly effective for e-commerce brands who want to bring static assets to life without a full video shoot.

For iterative projects, many users run a quick draft generation first to validate the scene composition, then refine the prompt with more specific motion instructions or camera angles before generating the final version. Seedance-20's speed makes this back-and-forth practical and cost-effective.

Use Cases

Seedance-20 is a strong fit for social media content creators who need a steady stream of short-form video clips. Platforms like Instagram Reels, TikTok, and YouTube Shorts reward frequent posting, and seedance-20 makes it possible to produce visually compelling content without a camera crew or editing suite.

Marketing and advertising teams use seedance-20 to rapidly prototype video concepts before committing to production budgets. A 5-second AI-generated clip can communicate a campaign idea to stakeholders far more effectively than a static mood board.

Game developers and concept artists use it to visualize environmental scenes, character animations, or cinematic cutscene ideas early in the design process. Educators and e-learning creators use seedance-20 to generate illustrative video segments that make abstract concepts more tangible for learners. The model's versatility across styles — realistic, stylized, abstract — means it adapts to a wide range of creative and professional contexts.