jianying-video-gen

Jianying AI Video Generator

Automate Jianying (xyq.jianying.com) via Playwright to generate AI videos using the Seedance 2.0 model.

Prerequisites

1. Python 3.9+ + playwright Installed
2. Chromium via playwright install chromium installed
3. cookies.json — Jianying login credentials (exported from browser), placed in the same directory as the script

pip install playwright && playwright install chromium

Core Script

scripts/jianying_worker.py — Main automation script

Usage

Text-to-Video (T2V)

python3 scripts/jianying_worker.py \
  --cookies /path/to/cookies.json \
  --output-dir /path/to/output \
  --prompt "赛博朋克风格的长安城，飞行汽车穿梭在霓虹灯笼之间" \
  --duration 10s \
  --model "Seedance 2.0"

Image-to-Video (I2V)

python3 scripts/jianying_worker.py \
  --cookies /path/to/cookies.json \
  --output-dir /path/to/output \
  --ref-image /path/to/image.png \
  --prompt "将这张图片变成动画，镜头从左向右缓慢平移" \
  --duration 10s \
  --model "Seedance 2.0 Fast"

Reference Video Generation (V2V)

python3 scripts/jianying_worker.py \
  --cookies /path/to/cookies.json \
  --output-dir /path/to/output \
  --ref-video /path/to/reference.mp4 \
  --prompt "画风改成宫崎骏风格，其他不变" \
  --duration 10s \
  --model "Seedance 2.0"

Extend

python3 scripts/jianying_worker.py \
  --cookies /path/to/cookies.json \
  --output-dir /path/to/output \
  --extend-url 'https://xyq.jianying.com/home?tab_name=integrated-agent&thread_id=YOUR_THREAD_ID&agent_name=pippit_video_part_agent' \
  --prompt "向后延伸，保持人物和镜头语言一致" \
  --duration 5s

Dry-Run Mode (for debugging)

python3 scripts/jianying_worker.py --cookies /path/to/cookies.json --prompt "测试" --dry-run

Only fill out the form without submitting, generate step_*.png screenshot for inspection. For V2V, dry-run now also outputs a 表单状态 JSON to confirm whether the reference video thumbnail is attached and whether the send button is unlocked.

Parameter Description

Parameter	Default Value	Optional Values	Description
--prompt	"A beautiful woman dancing"	Any text	Video description
--duration	10s	5s, 10s, 15s	Video duration
--ratio	横屏	横屏, 竖屏, 方屏	Aspect ratio
--model	Seedance 2.0	Seedance 2.0, Seedance 2.0 Fast	Model selection
--ref-image	None	Local image path	Reference image for I2V mode
--ref-video	None	Local video path	Reference video for V2V mode
--extend-url	None	thread link	Extend mode, continue generating based on an existing video
--cookies	cookies.json	File path	Jianying login credential path
--output-dir	.	Directory path	Output video save directory
--dry-run	false	-	Only fill out the form without submitting

Models and Credits

Model	Credits/Second	5s	10s	15s	Features
Seedance 2.0 Fast	3	15	30	45	Fast, suitable for testing
Seedance 2.0	5	25	50	75	High quality, for final production

Automation Workflow

登录(cookies) → 选模式(沉浸式短片) → 选模型 → [上传参考素材/打开已有 thread 并点向后延伸] → 选时长 → 输入Prompt → 发送
  → 拦截 thread_id → 导航详情页 → 轮询视频 → curl 下载 MP4

Prompt Writing Guide

For detailed prompt examples and writing tips, see references/prompt-guide.md。

FAQ

Q: What should I do if the cookies expire? Log in to xyq.jianying.com in the browser and export cookies.json using an extension such as EditThisCookie.

Q: Download 403? The script uses the thread_id detail page + curl download; the CDN link does not require cookies. If it still fails, check the network.

Q: Uploading the reference video is very slow? This is normal; an 8MB video takes about 60-90 seconds. The script will automatically wait up to 5 minutes.