Model checkpoints, reference distributions, and evaluation assets for MGFlow, presented in Unifying Distributional Training for One-Step Visual Generation. The release covers ImageNet class-conditional generation at 256×256 with JiT and pMF (B/L/H), and text-to-image generation at 512×512 with FLUX.2 [klein] 4B. Both generate images in one step after MGFlow post-training. The T2I checkpoints use the MGFlow-T2I reference dataset, which contains synthetic COCO and GenEval images paired with their prompts. The dataset link in the metadata applies to this T2I branch. If you use MGFlow or these assets, please cite
Independent publisher
Haoyang Shi
shy0423
Models in Library1
Datasets in Library1
Models on Hugging Face2
Followers—
Compare the published evidence
This profile connects the 2 model and dataset records shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Text to image.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.
Models
Datasets
Synthetic image-prompt pairs used as the reference distribution for MGFlow text-to-image post-training in The same reference images support both image-only and joint image-text matching. Model checkpoints and fitted reference distributions: The dataset contains 301,706 synthetic images at 512×512 with English prompts. All images were generated with the official distilled [FLUX.2 [klein] 4B](https://huggingface.co/black-forest-labs/FLUX.2-klein-4B) model using four sampling steps and guidance scale 1. Reference construction follows the procedure used by iRDM and AMFD: generate 24 candidates per caption, and retain the three with the highest PickScore. GenEval correctness test, considering up…