Updated: May 10, 2026
character
已按新格式处理：YouTube 短链已转完整版；RunningHub 已统一为 .ai；Bilibili 已清理为公开播放链接。解析依据为你上传的 PromptLine+Anima 艺术家批量风格测试.json 工作流。
https://www.youtube.com/watch?v=wdpemAhTCyM
This workflow is designed for PromptLine + Anima artist-style batch testing. Its main purpose is to let creators test many artist tags, style names, or prompt variations in sequence while keeping the main character prompt and generation settings consistent. Instead of manually changing the prompt one by one, this workflow uses a prompt-line structure to automatically feed different artist-style entries into the positive prompt, making it much more efficient for style research, model comparison, and Civitai showcase preparation.
The workflow uses Anima Preview as the main generation model, with anima-preview.safetensors loaded through UNETLoader. It uses qwen_3_06b_base.safetensors as the text encoder and qwen_image_vae.safetensors as the VAE. The generation canvas is set to 1024 x 1024 through EmptyLatentImage, which is a practical square format for artist-style comparison, anime character testing, preview images, and social media examples.
The key node in this workflow is easy promptLine. A list of artist-style tags is placed inside the promptLine node, with each line acting as a separate style entry. The workflow then combines that selected line with a fixed base prompt through JWStringConcat. This means the user can keep the same subject, pose, composition, and quality prompt, while only changing the artist-style direction. This is very useful when you want to compare how different names affect line quality, color treatment, facial design, background logic, lighting, and overall image mood.
The base prompt is stored in a String Literal node. It contains the shared generation description, including quality terms, character description, visual composition, lighting direction, and stylistic details. By separating the reusable base prompt from the artist-style list, the workflow becomes easier to manage. You can update the main subject once, then let promptLine test many styles automatically.
The first generation stage uses KSampler with Anima Preview, around 40 steps, CFG 4, er_sde sampler, simple scheduler, and full denoise. After that, the workflow includes a second Anima route using an additional LoRA, animaPreviewRdbt.4bB9.safetensors, with another KSampler pass. This second stage gives users another route for style enhancement or comparison, using different sampler settings such as Euler ancestral and normal scheduler. The workflow also applies AdjustContrast near the end to strengthen the final visual output.
This setup is useful for artists, prompt researchers, model testers, and workflow publishers who want to build a clean style matrix. It can help you quickly identify which artist tags produce stronger faces, better color harmony, cleaner linework, more dynamic lighting, or more distinctive anime aesthetics. It is also useful for preparing Civitai preview grids, RunningHub examples, YouTube thumbnail research, and batch testing before building a final production prompt.
In short, this is a practical PromptLine-powered Anima style research workflow. It turns repetitive manual prompt testing into a structured batch process. If you want to see how the artist list is connected, how the base prompt is combined with each style entry, and how the Anima output changes across different prompt lines, watch the full tutorial from the YouTube link above.
⚙️ Try the Workflow Online
👉 Workflow: https://www.runninghub.ai/post/2029160162027638786?inviteCode=rh-v1111
Open the link above to run the workflow directly online and view the generation results in real time.
If the results meet your expectations, you can also deploy it locally for further customization.
🎁 Fan Benefits: Register now to get 1000 points, plus 100 daily login points — enjoy 4090-level performance and 48 GB of powerful compute!
📺 Bilibili Updates (Mainland China & Asia-Pacific)
If you are in Mainland China or the Asia-Pacific region, you can watch the video below for workflow demos and a detailed creative breakdown.
📺 Bilibili Video: https://www.bilibili.com/video/BV15dPezmEaz/
I will continue updating model resources on Quark Drive:
👉 https://pan.quark.cn/s/20c6f6f8d87b
These resources are mainly prepared for local users, making creation and learning more convenient.
⚙️ 在线体验工作流
👉 工作流： https://www.runninghub.ai/post/2029160162027638786?inviteCode=rh-v1111
打开上方链接即可直接运行该工作流，实时查看生成效果。
如果觉得效果理想，你也可以在本地进行自定义部署。
🎁 粉丝福利： 注册即送 1000 积分，每日登录 100 积分，畅玩 4090 体验 48 G 超级性能！
📺 Bilibili 更新（中国大陆及南亚太地区）
如果你在中国大陆或南亚太地区，可以通过下方视频查看该工作流的实测效果与构思讲解。
📺 B站视频： https://www.bilibili.com/video/BV15dPezmEaz/
我会在 夸克网盘 持续更新模型资源：
👉 https://pan.quark.cn/s/20c6f6f8d87b
这些资源主要面向本地用户，方便进行创作与学习。