This workflow is designed for PromptLine + Anima Preview2 artist-style batch testing. Its main purpose is to help creators test many artist-style tags or style directions under the same base character prompt, same model, same resolution, and similar generation settings. Instead of manually changing the prompt again and again, this workflow uses a PromptLine structure to feed different style entries into the prompt pipeline, making it much faster to compare visual differences between multiple styles.
The workflow uses anima-preview2.safetensors as the main generation model, qwen_3_06b_base.safetensors as the text encoder, and qwen_image_vae.safetensors as the VAE. The canvas starts from a 1024 x 1024 empty latent, which is a practical square format for anime-style testing, Civitai preview images, RunningHub demos, style grids, and social media showcase images.
The key part of this workflow is the easy promptLine node. It contains a list of artist-style entries, with each line acting as one test direction. The selected line is combined with a fixed base prompt through JWStringConcat. This means the workflow can keep the core subject stable while changing only the style reference. In the uploaded setup, the base prompt describes a high-quality anime character: one girl, solo, full body, dynamic pose, short black bob hair, amber eyes, a beauty mark, silver cross earring, white military-style jacket, black pleated skirt, black gloves, and black thighhighs. By keeping this base prompt stable, the workflow makes style comparison much clearer.
The generation route uses Anima Preview2 with DPM++ 2M SDE GPU sampling and SGM Uniform scheduling. The first KSampler uses 30 steps, CFG 3, and full denoise to create the main image. After that, the workflow uses LatentUpscaleBy at 1.5x and sends the result into a second KSampler pass with 20 steps, CFG 2.5, and denoise around 0.65. This second pass acts as a refinement stage, improving detail, line quality, and visual polish without completely discarding the first result.
The negative prompt suppresses common problems such as low quality, blur, bad anatomy, bad hands, extra fingers, fused fingers, deformed faces, text, watermark, logo, and JPEG artifacts. This keeps the comparison cleaner and makes the differences between style entries easier to observe.
This workflow is especially useful for prompt researchers, anime-style testers, Civitai publishers, RunningHub workflow creators, and anyone who wants to build a reliable artist-style comparison set. It can help identify which prompt-line entries produce stronger faces, better color design, cleaner linework, more cinematic lighting, or more distinctive anime aesthetics. If you want to see how the PromptLine list, base prompt, two-stage generation, and final style comparison are connected, watch the full tutorial from the YouTube link above.
⚙️ Try the Workflow Online
👉 Workflow: https://www.runninghub.ai/post/2033542990416846849/?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/BV1Q1w1zKEwk/
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/2033542990416846849/?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/BV1Q1w1zKEwk/
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。