Watch the full video first if you want to understand how this FLUX.2 timestep distillation workflow works in practice. The video shows how FLUX.2 can run through a compact eight-step generation route, how optional reference images can be connected through latent conditioning, and how to launch the workflow online without rebuilding the full ComfyUI environment locally.
This ComfyUI workflow is designed for FLUX.2 timestep distillation image generation and reference-guided image editing. Its main purpose is to reduce the sampling cost of FLUX.2 while keeping the workflow clean, practical, and easy to test. Instead of using a heavy full-step FLUX.2 generation route, this workflow uses a distilled eight-step model path, making it more suitable for fast prompt testing, image editing experiments, reference-based generation, and online workflow demonstrations.
The workflow is built around flux2_distilled_8step_fp8mixed.safetensors as the main diffusion model. It uses mistral_3_small_flux2_fp8.safetensors as the Flux2 text encoder, flux2-vae.safetensors as the VAE, Flux2Scheduler for step scheduling, BasicGuider for the core guidance route, KSamplerSelect with Euler sampling, RandomNoise, EmptyFlux2LatentImage, SamplerCustomAdvanced, VAEDecode, and SaveImage. The whole graph is compact, which makes the logic easier to understand than a large multi-stage image workflow.
The most important idea is timestep distillation. The workflow uses an eight-step distilled FLUX.2 model, meaning the sampling process is compressed into a much smaller number of steps while still trying to preserve usable image quality, prompt following, composition, and detail. This is useful when creators need speed, repeated testing, or fast comparison between prompts and reference images.
The prompt side uses a structured JSON-style editing prompt. In the uploaded workflow, the example task focuses on single-image food editing: transforming a pasta dish into a Michelin-style lobster truffle tagliatelle while preserving the bowl angle, tabletop composition, shallow depth of field, and realistic food photography base. This shows that the workflow is not only for pure text-to-image generation, but also for controlled image editing and commercial-style visual transformation.
The reference image system is optional. The graph includes ReferenceLatent nodes and VAEEncode nodes. If the ReferenceLatent path is enabled, the workflow can use one or more reference images to guide the generated result. If the reference nodes are bypassed, the workflow becomes a cleaner text-to-image generation setup. This makes the workflow flexible: creators can use it for pure prompt generation, single-image editing, or reference-guided visual control.
The image size is handled through the input image and EmptyFlux2LatentImage. The workflow also includes image scaling tools, allowing the reference image to be resized before being encoded into the latent path. The final result is decoded through Flux2 VAE and saved as an image output.
Compared with ordinary FLUX.2 workflows, this version is faster and more direct. It is suitable for commercial food retouching, product image editing, reference-based composition control, prompt testing, visual replacement, concept exploration, RunningHub demonstrations, YouTube tutorials, Bilibili showcases, and Civitai workflow publishing.
Main features:
FLUX.2 timestep distillation workflow
Eight-step distilled FLUX.2 generation route
flux2_distilled_8step_fp8mixed model support
Mistral Flux2 fp8 text encoder
Flux2 VAE decoding
Flux2Scheduler step control
Euler sampler with SamplerCustomAdvanced
BasicGuider clean guidance route
Optional ReferenceLatent image guidance
Can work as text-to-image when reference nodes are bypassed
Supports single-image editing logic
Structured JSON-style editing prompt
Good for fast testing and commercial image transformation
Final SaveImage output
Suggested workflow:
Start with a clear editing or generation goal. If you want pure text-to-image generation, keep the ReferenceLatent nodes bypassed and write a strong prompt describing subject, composition, material, lighting, camera style, and final use case. If you want image editing or reference-guided generation, enable the ReferenceLatent path and load a clean reference image. Keep the reference image composition simple and readable. Use the eight-step distilled route for fast testing first. If the composition is wrong, adjust the prompt before changing advanced settings. If the reference image is too dominant, reduce reliance on reference conditioning. If the result does not preserve enough structure, make the preservation rules more explicit in the prompt.
⚙️ RunningHub Workflow
Try the workflow online right now — no installation required.
👉 Workflow: https://www.runninghub.ai/post/2061782953197391873?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/BV1yLEc6dEJc/
☕ 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/2061782953197391873?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/BV1yLEc6dEJc/
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。