FlowUpscaler
Flow Upscaler is a fast latent upscaler model that works in the Flux.2 latent space.
Source du modèle
- Type
- Other
- Taille du fichier
- 226 MB
- Modèle de base
- black-forest-labs/FLUX.2-klein-4B
- Version
- vRevision ae33429
- Mis à jour
- 9 oct. 2026
- Téléchargements
- 46
- Source
- Hugging Face
Description de la source
Flow Upscaler
Flow Upscaler is a fast latent upscaler model that works in the Flux.2 latent space.
Under the hood, it is a lightweight Rectified Flow model with 59M parameters that generates upscaled latents in a single denoising step.
Features:
- Upscaling from 512x512 to 1024x1024 takes 8ms*
- The model is trained for 2X upscaling, but multiple passes can be chained to reach up to 8K resolution
- A full pipeline with Flux generation, upscaling to 8K, and decoding runs in just 25 seconds (on RTX 5090)
- The training process uses Flow Distillation with Flux.2 as a teacher, forcing the model to learn strong image semantics
*On RTX 5090, in latent space, without decoding, see benchmark here.
Here is one 4X upscaled image (two passes):
comparison
How it works
Architecturally, Flow Upscaler is a U-Net with SDXL-style ResNet blocks. It takes a noisy sample as input and predicts velocity as output. The generation process happens directly in high-resolution latent space.
The low-resolution latents are passed through a separate conditioning encoder that produces control signals, which are injected into the main U-Net encoder using FiLM conditioning.
No attention layers are used, so compute scales linearly with image area. This makes generation at 8K resolution possible.
example
The model is trained using Flow Distillation with Flux.2-klein-4B as a teacher. We generated 20K diverse images with Flux, storing the initial noise, generated latents, and downscaled latents used for conditioning.
The downscaled latents are created by decoding high-resolution latents, downscaling them in pixel space, and encoding them back into latents. Direct latent downscaling introduces artifacts and breaks latent patterns, resulting in blurry decoded images.
example
Training code
If you want to explore the training code or use the model outside ComfyUI, see:
notebooks/flow_upscaler in https://github.com/tensorforger/CTGMWorkshop
Sources
1 source
Sources
1 sourceVérifié 9 oct.
Artefacts du modèle
1 artefact
Artefacts du modèle
1 artefactflow_upscaler.safetensors
safetensors · 226 MB · SHA-256 91eb93b40179…1839 · Hugging Face
Download fileExtraits de sources
3 extraits
Extraits de sources
3 extraits--- license: unlicense tags: - flux2 - flux - image-to-image - image-upscaling - latent-upscaling - super-resolution - diffusion - flow-matching - rectified-flow - generative-models - image-generation - pytorch library_name: pytorch base_model: - black-forest-labs/FLUX.2-klein-4B pipeline_tag: image-to-image --- # Flow Upscaler **Flow Upscaler** is a fast latent upscaler model that works in the [Flux.2](https://bfl.ai/models/flux-2) latent space. Under the hood, it is a lightweight **Rectified Flow** model with **59M** parameters that generates upscaled latents in a single denoising step. **[ComfyUI Node](https://github.com/TensorForger/comfyui-flow-upscaler)** Features: * Upscaling from **512x512** to **1024x1024** takes **8ms*** * The model is trained for **2X** upscaling, but multiple passes can be chained to reach up to **8K** resolution * A full pipeline with Flux generation, upscaling to **8K**, and decoding runs in just **25 seconds** (on RTX 5090) * The training process uses **Flow Distillation** with Flux.2 as a teacher, forcing the model to learn strong image semantics *On RTX 5090, in latent space, without decoding, see benchmark [here](https://github.com/tensorforger/CTGMWorkshop). Here is one **4X** upscaled image (two passes):  ## How it works Architecturally, Flow Upscaler is a U-Net with SDXL-style ResNet blocks. It takes a noisy sample as input and predicts velocity as output. The generation process happens directly in high-resolution latent space. The low-resolution latents are passed through a separate conditioning encoder that produces control signals, which are injected into the main U-Net encoder using FiLM conditioning. No attention layers are used, so compute scales linearly with image area. This makes generation at **8K** resolution possible.  The model is trained using **Flow Distillation** with Flux.2-klein-4B as a teacher. We generated **20K** diverse images with Flux, storing the initial noise, generated latents, and downscaled latents used for conditioning. The downscaled latents are created by decoding high-resolution latents, downscaling them in pixel space, and encoding them...
Source context: 46 downloads · 20 likes · Pipeline image-to-image · Library pytorch · Repo TensorForger/FlowUpscaler
Source context: 53 downloads · 15 likes · Pipeline image-to-image · Library pytorch · Repo TensorForger/FlowUpscaler
