Category: provider
Model Studio Aishi Video Generation
Validation
CODEBLOCK0
Pass criteria: command exits 0 and output/aliyun-pixverse-generation/validate.txt is generated.
Output And Evidence
- Save normalized request payloads, chosen model variant, and task polling snapshots under output/aliyun-pixverse-generation/. Record region, resolution/size, duration, and whether audio generation was enabled.
Use Aishi when the user explicitly wants the non-Wan PixVerse family for video generation.
Critical model names
Use one of these exact model strings:
- INLINECODE2 INLINECODE3 INLINECODE4 INLINECODE5
Selection guidance:
- Use pixverse/pixverse-v5.6-t2v for text-only generation. Use pixverse/pixverse-v5.6-it2v for first-frame image-to-video. Use pixverse/pixverse-v5.6-kf2v for first-frame + last-frame transitions. Use pixverse/pixverse-v5.6-r2v for multi-image character/style consistency.
Prerequisites
- This family currently only supports China mainland (Beijing). Install SDK or call HTTP directly:
CODEBLOCK1
- Set DASHSCOPE_API_KEY in your environment, or add dashscope_api_key to ~/.alibabacloud/credentials.
Normalized interface (video.generate)
Request
- model (string, required) INLINECODE14 (string, optional for it2v, required for other variants) INLINECODE16 (array, optional) INLINECODE17 (string, optional): direct pixel size such as 1280*720, used by t2v and INLINECODE20 INLINECODE21 (string, optional): 360P/540P/720P/1080P, used by it2v and INLINECODE27 INLINECODE28 (int, required): 5/8/10, except 1080P only supports 5/ INLINECODE33 INLINECODE34 (bool, optional) INLINECODE35 (bool, optional) INLINECODE36 (int, optional)
Response
- task_id (string) INLINECODE38 (string) INLINECODE39 (string, when finished)
Endpoint and execution model
- Submit task: INLINECODE40 Poll task: INLINECODE41 HTTP calls are async only and must set header X-DashScope-Async: enable.
Quick start
Text-to-video:
CODEBLOCK2
Image-to-video:
CODEBLOCK3
Operational guidance
- t2v and r2v use size; it2v and kf2v use resolution. For kf2v, provide exactly one first_frame and one last_frame. For r2v, you can pass up to 7 reference images. Aishi returns task IDs first; do not treat the initial response as the final video result.
Output location
- Default output: INLINECODE53 Override base dir with OUTPUT_DIR.
References
技能名称: aliyun-pixverse-generation
详细描述:
类别: provider
Model Studio Aishi 视频生成
验证
bash
mkdir -p output/aliyun-pixverse-generation
python -m pycompile skills/ai/video/aliyun-pixverse-generation/scripts/prepare aishirequest.py && echo py compile_ok > output/aliyun-pixverse-generation/validate.txt
通过标准:命令退出码为0,且已生成 output/aliyun-pixverse-generation/validate.txt。
输出与证据
- 将标准化请求负载、所选模型变体及任务轮询快照保存至 output/aliyun-pixverse-generation/ 目录下。 记录区域、分辨率/尺寸、时长以及是否启用音频生成。
当用户明确希望使用非Wan系列的PixVerse系列进行视频生成时,使用Aishi。
关键模型名称
使用以下精确的模型字符串之一:
- pixverse/pixverse-v5.6-t2v pixverse/pixverse-v5.6-it2v pixverse/pixverse-v5.6-kf2v pixverse/pixverse-v5.6-r2v
选择指南:
- 纯文本生成使用 pixverse/pixverse-v5.6-t2v。 首帧图生视频使用 pixverse/pixverse-v5.6-it2v。 首帧+末帧过渡使用 pixverse/pixverse-v5.6-kf2v。 多图像角色/风格一致性使用 pixverse/pixverse-v5.6-r2v。
前提条件
- 该系列目前仅支持中国大陆(北京)区域。 安装SDK或直接调用HTTP:
bash
python3 -m venv .venv
. .venv/bin/activate
python -m pip install dashscope
- 在环境中设置 DASHSCOPEAPI KEY,或将其添加到 ~/.alibabacloud/credentials 文件中。
标准化接口 (video.generate)
请求
- model (字符串,必填) prompt (字符串,it2v 可选,其他变体必填) media (数组<对象>,可选) size (字符串,可选):直接像素尺寸,如 1280*720,用于 t2v 和 r2v resolution (字符串,可选):360P/540P/720P/1080P,用于 it2v 和 kf2v duration (整数,必填):5/8/10,但1080P仅支持 5/8 audio (布尔值,可选) watermark (布尔值,可选) seed (整数,可选)
响应
- taskid (字符串) taskstatus (字符串) video_url (字符串,任务完成时)
端点与执行模型
- 提交任务:POST https://dashscope.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis 轮询任务:GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} HTTP调用仅支持异步模式,必须设置请求头 X-DashScope-Async: enable。
快速开始
文本生成视频:
bash
python skills/ai/video/aliyun-pixverse-generation/scripts/prepareaishi request.py \
--model pixverse/pixverse-v5.6-t2v \
--prompt 一个紧凑型机器人在霓虹闪烁的雨巷中行走。 \
--size 1280*720 \
--duration 5
图像生成视频:
bash
python skills/ai/video/aliyun-pixverse-generation/scripts/prepareaishi request.py \
--model pixverse/pixverse-v5.6-it2v \
--prompt 乌龟缓慢游动,镜头随之上升。 \
--media image_url=https://example.com/turtle.webp \
--resolution 720P \
--duration 5
操作指南
- t2v 和 r2v 使用 size;it2v 和 kf2v 使用 resolution。 对于 kf2v,需提供恰好一个 firstframe 和一个 last frame。 对于 r2v,最多可传入7张参考图像。 Aishi 首先返回任务ID;请勿将初始响应视为最终视频结果。
输出位置
- 默认输出:output/aliyun-pixverse-generation/request.json 可通过 OUTPUT_DIR 覆盖基础目录。
参考资料