Watch the full video first if you want to understand how this Boogu-Image 0.1 Turbo text-to-image workflow works in practice. The video shows how a simple text prompt can be turned into a finished image through a very compact and fast generation route.
This ComfyUI workflow is designed for Boogu-Image 0.1 Turbo text-to-image generation. Its main purpose is speed. Instead of using a large multi-stage rendering graph, a complex reference system, or a heavy post-processing chain, this workflow keeps the entire generation route clean and direct: load the Turbo model, encode the text prompt, create a latent canvas, sample the image quickly, decode the result, preview it, and save the final output.
The workflow is built around boogu_image_turbo_bf16.safetensors as the main image generation model. This is the core Turbo model used for fast text-to-image output. The text encoder uses qwen3vl_8b_fp8_scaled.safetensors with the Boogu type setting, allowing the workflow to understand the text prompt and convert it into conditioning for image generation. The VAE route uses ae.safetensors to decode the final latent result into a visible image.
The prompt section is intentionally simple. A standard CLIPTextEncode node receives the user prompt and creates the positive conditioning. In the sample workflow, the prompt describes a red panda holding a sign that says “BOOGU!”. The negative route is handled by ConditioningZeroOut, which creates a clean empty negative conditioning from the positive branch. This keeps the workflow lightweight and avoids requiring users to manage a separate negative prompt field for basic generation.
The latent canvas is created by EmptyLatentImage at 1024×1024 with batch size 1. This makes the workflow suitable for square image generation, AI app thumbnails, quick concept tests, product-style experiments, character ideas, posters, icons, stickers, and fast creative drafts. Users can later adjust the width, height, or batch size if they want a different output format.
The sampling stage uses a KSampler with only 4 steps, CFG 1, LCM sampler, SGM Uniform scheduler, fixed seed, and denoise set to 1. This is the main reason the workflow feels fast. It is configured for quick generation rather than slow, repair-heavy refinement. The low step count makes it useful for rapid testing, prompt exploration, and high-volume creative iteration.
After sampling, VAEDecode converts the latent into the final image. PreviewImage gives an immediate visual result inside the workflow, while SaveImage exports the final image with a standard filename prefix. The full graph is small, readable, and easy to reuse as a base module for a RunningHub AI application.
This workflow is suitable when you want a fast “type prompt → generate image” experience. It is best used for quick creative exploration, prompt testing, simple visual concepts, social media graphics, stylized characters, product ideas, and fast image drafts.
Main features:
Boogu-Image 0.1 Turbo text-to-image workflow
Fast prompt-to-image generation
boogu_image_turbo_bf16.safetensors main model
qwen3vl_8b_fp8_scaled.safetensors Boogu encoder
ae.safetensors VAE decoding
Simple CLIPTextEncode prompt route
ConditioningZeroOut empty negative conditioning
1024×1024 default latent canvas
Batch size 1 default output
KSampler generation route
4-step Turbo sampling
CFG 1 setting
LCM sampler
SGM Uniform scheduler
Denoise 1 full generation
PreviewImage for quick checking
SaveImage for final export
Clean and reusable text-to-image structure
Suggested workflow:
Write one clear prompt that describes the subject, style, composition, lighting, and key visual details. Because this is a Turbo workflow, keep the prompt direct and avoid stacking too many conflicting instructions. Start with the default 1024×1024 canvas and 4-step setting for fast tests. If the result is not specific enough, strengthen the subject description and add clearer composition details. If the image becomes messy, simplify the prompt and remove unnecessary style conflicts. Use this workflow as a fast first-pass generator before moving promising results into a slower refinement, editing, or upscaling workflow.
⚙️ RunningHub Workflow
Try the workflow online right now — no installation required.
👉 Workflow: https://www.runninghub.ai/post/2067853157933338626?inviteCode=rh-v1111
If the results meet your expectations, you can later deploy it locally for customization.
🎁 Fan Benefits: Register to get 1000 points + daily login 100 points — enjoy 4090 performance and 48 GB super power!
📺 Bilibili Updates (Mainland China & Asia-Pacific)
If you’re in the Asia-Pacific region, you can watch the video below to see the workflow demonstration and creative breakdown.
📺 Bilibili Video: https://www.bilibili.com/video/BV17ujH6eEZk/
☕ Support Me on Ko-fi
If you find my content helpful and want to support future creations, you can buy me a coffee ☕.
Every bit of support helps me keep creating — just like a spark that can ignite a blazing flame.
👉 Ko-fi: https://ko-fi.com/aiksk
💼 Business Contact
For collaboration or inquiries, please contact aiksk95 on WeChat.
⚙️打开下方链接即可在线体验，无需安装。
👉 工作流： https://www.runninghub.ai/post/2067853157933338626?inviteCode=rh-v1111
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/BV17ujH6eEZk/
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。